System and method of disambiguating natural language processing requests

Description

FIELD OF THE INVENTION

The invention relates to a system and method of disambiguating natural language processing requests based on smart matching, request confirmations that are used until ambiguities are resolved, and machine learning.

BACKGROUND OF THE INVENTION

Facilitating human to machine interactions in a natural manner is a difficult problem. Both non-voice and certain voice interfaces tend to be overly structured and require some level of familiarity with the interfaces. An ability for a human to interact with machines using natural language utterances in various contexts remains a desirable goal. However, human interaction with machines in various contexts remains a difficult problem.

One area of difficulty lies in the disambiguation of natural language requests. The objects of natural language requests and command, entity names, may sometimes be ambiguous. For example, a user may have multiple contacts of the same name in their directory. A phone dialing request made by first name may require disambiguation. A natural language system however, that constantly requests a user to use specific and unambiguous language or to clarify requests made in ambiguous language may quickly begin to feel burdensome for the user.

These and other drawbacks exist with conventional natural language processing disambiguation methods.

SUMMARY OF THE INVENTION

The invention addressing these and other drawbacks relates to a system and method of machine learned disambiguation of natural language processing requests. An intent of a natural language input (e.g., a natural language utterance of a user) may be ambiguous. For example, a user may provide the input “call John,” when there are several “Johns” in the user's contact list. Other types of ambiguities and other types of requests may be ambiguous based on natural language processing as well.

The system may disambiguate natural language processing requests based on smart matching, request confirmations that are used until ambiguities are resolved, and machine learning. Smart matching may match entities (e.g., contact names, place names, etc.) based on user information such as call logs, user preferences, etc. If multiple matches are found and disambiguation has not yet been learned by the system for these matches, the system may request that the user identify the intended entity. On the other hand, if disambiguation has been learned by the system, the system may execute the request without confirmations. The system may use a record of confirmations and/or other information to continuously learn a user's inputs in order to reduce ambiguities and no longer prompt for confirmations.

To this end, the system may include a smart matching engine that identifies and matches entities, a confirmation engine that generates request confirmations (e.g., “do you want to call John Doe or John Smith”?) that are used until a sufficient level of confidence that an ambiguity relating to a given natural language input has been resolved (e.g., until the system is confident that the user intends to call “John Doe” and not “John Smith”), a learning engine that uses machine learning based on user information (e.g., history of request confirmations from a user, where each request confirmation sought to resolve a prior ambiguous result of natural language processing from a prior natural language input), and/or other components to disambiguate a natural language input.

The smart matching engine may identify an input entity name. The smart matching engine may then match the input entity name to an entity name that can be acted upon. For example, if the user requests directions to Main St., then the smart matching engine may operate to suggest a match between the user uttered input entity name “Main St.” and an identified entity name—i.e., an actual, mappable, Main St. If the user asks to “call James,” then the smart matching engine may operate to suggest a match between an identified entity name, i.e., an actual contact in the user's directory, and the input entity name of “James.”

The smart matching engine may suggest a match between an input entity name and an identified entity name based at least partially on an entity matching model. The entity matching model may comprise information compiled through machine learning techniques, and may include at least context information and user history information, as well as other information suitable for machine learning. A suggested match may include the input entity name and an identified entity name. A suggested match may further include additional entity match information.

Additional entity match information may include context information related to the natural language request, the input entity name, and the identified entity name. For example, for a user that lives in New York, an entity matching model may associate the entity name “taxi” in the request “call a taxi” with a particular taxi company in New York. When creating a suggested match for the input entity name “taxi,” for that user, the smart matching engine may include the additional entity match information of the user's location as part of the suggested match. This may permit the system to disambiguate the term “taxi” differently based on the user's location.

Context information usable by the smart matching engine may include a request context. For example, if a user makes a request to “call James” during the business day, the smart matching engine may suggest a match between the input entity name James and a business contact named James. If the request is made after hours, the smart matching engine may suggest a match with a contact from a friends and family contact list. Similarly, when attempting to disambiguate requests for directions, smart matching engine may suggest matches between locations that are closer to a present location of the user.

User history information usable by the smart matching engine may include information about previous user requests and activity. For example, when requesting to make a call, the smart matching engine may select a name from a user's contact list that the user calls very frequently. When seeking directions, the smart matching engine may suggest a destination location that matches a location the user has travelled to frequently.

The smart matching engine may additionally employ machine learned matching techniques. For example, a user may frequently refer to a contact name by a nickname, making a request to “call Jim,” where the name “James” is stored in the user's contact list. Such a connection may become a part of an entity matching model based on machine learning techniques, as described in more detail below.

In some implementations, the smart matching engine may suggest a match between an input entity name and an identified entity name based on an exact match. For example, where a user requests to “call John,” the system may suggest only matches that match “John” exactly. In some implementations, the smart matching engine may suggest a match between an input entity name and an identified entity name based on a phonetic match. For example, where a user requests to “call John,” the system may suggest all matches that sound like “John,” including, for example, “Jon.” In some implementations, the smart matching engine may suggest a match between an input entity name and an identified entity name based on similar sounds. Continuing to use the above example, the smart matching engine may suggest a match between “John” and “Juwan.” In some implementations the smart matching engine may suggest a match between an input entity name and an identified entity name based on an interpretation of a user's pronunciation style.

In some implementations, the smart matching engine may suggest a match between an input entity name and an identified entity name based on learned associations. For example, a user may request to “call my brother.” The smart matching engine may suggest a match between the input entity name “my brother,” and all user contacts having a same last name as the user. The smart matching engine may also request clarification from the user about the entity name “my brother.” Once the system has successfully confirmed a match between a user contact and the input entity name “my brother,” the entity matching model may be updated with the information in order to correctly link the input entity name “my brother” to the appropriate contact. In another example, a user may request directions to “the pizza place.” Smart matching engine may propose a suggested match between the input entity name “the pizza place,” and the closest pizza restaurant. Once the system has successfully confirmed a match between “the pizza place,” and a particular pizza restaurant, the entity matching model may be updated to associate “the pizza place,” with the specific pizza restaurant.

In some implementations, the smart matching engine may pass a suggested match between an input named entity and an identified named entity to confirmation engine for command confirmation, as described below.

In some implementations, the smart matching engine may determine an alternate identified entity name if the confirmation engine is unable to confirm the suggested match between the input named entity and the identified entity name. In some implementations, the smart matching engine may determine an alternate identified entity name using similar methods to those described above for selecting the first identified entity name. In determining an alternate identified entity name, the smart matching engine may select a next best match for an input entity name. In some implementations, the smart matching engine may develop an n-best list of suggested matches for an input entity name including n number of potential matches ranked in order of best fit. The smart matching engine may provide the 2nd, 3rd, 4th, etc., matches from an n-best list, in order, to the confirmation engine as alternate identified names if the first suggestion is cancelled or met with non-confirmation. In some implementations, the smart matching engine may provide an entire n-best list to the confirmation engine.

The confirmation engine may perform command confirmation based on a user's confirmation history of a proposed entity match. Command confirmation may be performed actively and/or passively. For example, active command confirmation may include querying the user to confirm a suggest match—e.g., where a user has asked to “call James,” a command confirmation may query the user “did you mean James Smith?” In some implementations, passive command confirmation may include permitting a user a predefined amount of time to cancel an action. For example, where a user has asked to “call James,” a command confirmation may alert the user “calling James Smith,” and wait for a predetermined amount of time to give the user a chance to cancel the action before executing the action. If the user does not cancel the action, then the suggested entity match may be confirmed. A predetermined amount of time may be relatively short, e.g., 1 to 5 seconds, or relatively long, e.g., 5-15 seconds. Selecting which command confirmation method to use may be based on a user's confirmation history.

For example, the confirmation engine may use a history of a user's confirmation between a particular input entity name and a particular identified entity name. For example, a user may ask to “call James,” and the smart matching engine may determine a suggested match between the input entity name “James” and an identified entity name “James Smith,” corresponding to a user contact. If the smart matching engine has previously suggested the same match, and the user has previously provided confirmations, the confirmation engine may alert the user of the impending action. If the user has not previously provided enough confirmations of the suggested match, the confirmation engine may request active confirmation from the user. The confirmation engine may switch between these two methods when a number of previous confirmations of the suggested match exceeds a predetermined threshold number of times. Such a predetermined threshold number may be one or more times. A predetermined threshold number may vary depending on the type of entity name that is being matched (e.g., whether it is a contact name or a location address.)

In some implementations, the confirmation engine may update a number of previous confirmations of a suggested match in a user's confirmation history after confirmation or non-confirmation is received from the user. If confirmation is received from the user for a particular suggested match, the confirmation engine may increment the number of previous confirmations. If a user cancels or fails to confirm a suggested match, the confirmation engine may instead decrement the number of previous confirmations. In some implementations, the confirmation engine may reset a number of previous confirmations to zero after a non-confirmation is received. In some implementations, the confirmation engine may separately record a number of non-confirmations and decrement or reset the previous confirmation score after a threshold number of non-confirmations is reached. In some implementations, the confirmation engine may separately record a number of consecutive non-confirmations and decrement or reset the previous confirmation score after a threshold number of consecutive non-confirmations is reached. In some implementations, the confirmation engine may do nothing to a previous confirmation score when a non-confirmation is received.

In some implementations, the confirmation engine may request an alternate identified match from the smart matching engine if a non-confirmation or cancellation is received. The smart matching engine may provide a next best alternate identified match to the confirmation engine. The confirmation engine may treat the alternate identified match similarly to the initial identified match, and determine an active or passive confirmation method based on a number of previous confirmations.

In some implementations, the confirmation engine may receive an n-best list from the smart matching engine. The confirmation engine may then try to confirm with the user each entry in the n-best list as a selected match. In some implementations, the confirmation engine may try each member of the n-best list in turn, using the above-discussed confirmation methods based on confirmation history. In some implementations, the confirmation engine may try to confirm the first entry of a an n-best list, and, if confirmation fails, query the user to select from the remaining entries in the n-best list.

The learning engine may provide machine learning updates to a smart matching model used by the smart matching engine and confirmation engine in suggesting and confirming matches between natural language inputs and named entities, and/or other instructions that computer system 110 to perform various operations, each of which are described in greater detail herein.

The learning engine may include instructions for updating an entity matching model based on results provided by the smart matching engine and the confirmation engine. As described above, the smart matching engine may suggest a match between an input entity name and an identified entity name, while the confirmation engine may operate to confirm the identified entity name and determine it as a selected entity name. When an entity name is confirmed as a selected entity name, the learning engine may update the entity matching model with information about the confirmed match. The learning engine may update context information of the entity matching model with information about the context (e.g., time, location, whether the user is traveling, etc.) under which the suggested match was confirmed. The learning engine may update user history information of the entity matching model, for example, to keep track of a frequency of user requests involving the selected entity name.

For example, where a user has a requested to make a phone call or send a text message to a contact and the confirmation engine confirms the suggested match provided by the smart matching engine, the learning engine may update the entity matching model with information about the match between the input entity name and the identified entity name, including, for example, time of day, location of the user, category of the entity name (e.g., business or personal contact), and any other suitable information. In some implementations, when the confirmation engine obtains a non-confirm result of a suggested match, the learning engine may update entity matching model with information that the match was not confirmed, as well as other context information associated with the failed match.

For example, a user may frequently his brother “James Lewis” in the evening but call his business associate “James Smith” during the day. Due to a frequency of calls, the smart matching engine may initially suggest a match between the request to “call James,” and “James Smith,” even when the call is taking place after 9 pm. After one or more non-confirmations between a command to “call James” at 9 pm and “James Smith,” the learning engine may sufficiently update the entity matching model to make an association between “James” and the user's brother “James Lewis” in the evening, and between “James” and the user's business associate “James Smith” during the day.

These and other objects, features, and characteristics of the system and/or method disclosed herein, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate corresponding parts in the various figures. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. As used in the specification and in the claims, the singular form of “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a system for natural language processing, according to an implementation of the invention.

FIG. 2 depicts an NLP system configured to perform natural language processing on an utterance, according to an implementation of the invention.

FIG. 3 depicts a process of natural language processing disambiguation, according to an implementation of the invention.

FIG. 4 depicts a process of disambiguation machine learning, according to an implementation of the invention.

DETAILED DESCRIPTION OF THE INVENTION

The invention described herein relates to a system and method of natural language processing disambiguation. In particular, the systems and methods described are related to machine learning of disambiguation models. The system may be configured to receive and process natural language inputs, and then use machine learning techniques to optimally provide disambiguation between potentially ambiguous inputs.

For example, a user may have multiple contacts having the same first or last name in a directory. If a user makes a natural language request to “Call John,” when there are three Johns stored in the users contact directory, the system may employ disambiguation models to determine which John the user the intended. The system may review context data, including call frequency, call timing, user location, and other information, to determine a suggestion to match the user's request. The user may then be prompted to confirm the suggestion, and/or, in some implementations, be provided with a time in which to cancel the suggestion, before the request is carried out using the suggestion. The system may incorporate the user response in disambiguation models to optimize future suggestions and actions. Further examples of user requests requiring disambiguation may include requests for directions (e.g., where multiple destinations may have similar names), requests for weather reports (e.g., where the user fails to specify the location of request), requests to play music (e.g., where the user selects an album but not an artist) and others. Disambiguation may be required between identical user requests, between phonetically similar requests, between homophonic requests, and/or between requests that may sound similar based on user pronunciation. The system may further be configured for disambiguation of generic requests to identify a user's specific intent. For example, a request to “call the pizza place,” may be disambiguated to a phone dialing request to contact a specific pizza restaurant that a user frequently calls.

FIG. 1 illustrates a system 100 for natural language processing. In one implementation, system 100 may include a computer system 110, a user device 140, a database 104, and/or other components. Components of system 100 may communicate with one another via network 102. Network 102 may be any type of network, such as a WAN, LAN, Bluetooth network, the Internet, etc.

Computer system 110 may be configured as a server (e.g., having one or more server blades, processors, etc.), a gaming console, a handheld gaming device, a personal computer (e.g., a desktop computer, a laptop computer, etc.), a smartphone, a tablet computing device, and/or other device that can be programmed for natural language processing, and, in particular the disambiguation of natural language processing requests.

Computer system 110 may include one or more processors 112 (also interchangeably referred to herein as processors 112, processor(s) 112, or processor 112 for convenience), one or more storage devices 114, and/or other components. Processors 112 may be programmed by one or more computer program instructions. For example, processors 112 may be programmed by NLP subsystem 106A, and/or other instructions that program computer system 110 to perform various operations, each of which are described in greater detail herein. As used herein, for convenience, the various instructions and engines will be described as performing an operation, when, in fact, the various instructions program the processors 112 (and therefore computer system 110) to perform the operation. Storage device 114 may further include a user history database 180.

NLP subsystem 106A may perform at least some aspects of natural language processing on an utterance using all or a portion of the components of NLP subsystem 106 illustrated in FIG. 2. In other words, computer system 110 may be programmed with some or all of the functions of NLP system 106 described with respect to FIG. 2.

User device 140 may be configured as a server device, a gaming console, a handheld gaming device, a personal computer (e.g., a desktop computer, a laptop computer, etc.), a smartphone, a tablet computing device, and/or other device that can be programmed to receive and interpret natural language processing requests. In particular, user device 140 may be programmed to perform disambiguation on a received natural language input.

User device 140 may include one or more processors 142 (also interchangeably referred to herein as processors 142, processor(s) 142, or processor 142 for convenience), one or more storage devices 144, and/or other components. Processors 142 may be programmed by one or more computer program instructions. For example, processors 142 may be programmed by NLP subsystem 106B, and/or other instructions that program computer system 110 to perform various operations, each of which are described in greater detail herein. As used herein, for convenience, the various instructions will be described as performing an operation, when, in fact, the various instructions program the processors 142 (and therefore user device 140) to perform the operation. Additionally storage device 144 may include one or more databases, including a user information a user history database 190.

Any or all aspects of natural language processing may be carried out by programming instructions instantiated on either NLP subsystem 106b of user device 140 and NLP subsystem 106a of computer system 110. Each of user device 140 and computer system 110 may perform all or a portion of a natural language processing task. User device 140 and computer system 110 may operate to perform natural language processing tasks sequentially, communicating results via network 102. In such implementations, one system may pass results of a completed task to the other system for further processing. User device 140 and computer system may operate to redundantly perform natural language processing tasks. In such implementations, depending upon the results of a natural language processing task carried out by one system, the other system may repeat the same task. User device 140 and computer system 110 may operate in parallel to perform natural language processing tasks. In such implementations, user device 140 and computer system 110 may each perform one or more tasks independently and, after task completion, communicate with each other to select between and or combine the processing results.

NLP subsystem 106B may perform at least some aspects of natural language processing on an utterance using all or a portion of the components of NLP subsystem 106 illustrated in FIG. 2. In other words, user device 140 may be programmed with some or all of the functions of NLP system 106 described with respect to FIG. 2.

Although illustrated in FIG. 1 as a single component, computer system 110 and user device 140 may each include a plurality of individual components (e.g., computer devices) each programmed with at least some of the functions described herein. In this manner, some components of computer system 110 and/or user device 140 may perform some functions while other components may perform other functions, as would be appreciated. The one or more processors 112 may each include one or more physical processors that are programmed by computer program instructions. The various instructions described herein are exemplary only. Other configurations and numbers of instructions may be used, so long as the processor(s) 112 are programmed to perform the functions described herein.

Furthermore, it should be appreciated that although the various instructions are illustrated in FIG. 1 as being co-located within a single processing unit, in implementations in which processor(s) 112 includes multiple processing units, one or more instructions may be executed remotely from the other instructions.

The description of the functionality provided by the different instructions described herein is for illustrative purposes, and is not intended to be limiting, as any of instructions may provide more or less functionality than is described. For example, one or more of the instructions may be eliminated, and some or all of its functionality may be provided by other ones of the instructions. As another example, processor(s) 112 may be programmed by one or more additional instructions that may perform some or all of the functionality attributed herein to one of the instructions.

The various instructions described herein may be stored in a storage device 114, which may comprise random access memory (RAM), read only memory (ROM), and/or other memory. The storage device may store the computer program instructions (e.g., the aforementioned instructions) to be executed by processor 112 as well as data that may be manipulated by processor 112. The storage device may comprise floppy disks, hard disks, optical disks, tapes, or other storage media for storing computer-executable instructions and/or data.

The various components illustrated in FIG. 1 may be coupled to at least one other component via a network 102, which may include any one or more of, for instance, the Internet, an intranet, a PAN (Personal Area Network), a LAN (Local Area Network), a WAN (Wide Area Network), a SAN (Storage Area Network), a MAN (Metropolitan Area Network), a wireless network, a cellular communications network, a Public Switched Telephone Network, and/or other network. In FIG. 1, as well as in other drawing Figures, different numbers of entities than those depicted may be used. Furthermore, according to various implementations, the components described herein may be implemented in hardware and/or software that configure hardware.

The various databases 104 described herein may be, include, or interface to, for example, an Oracle™ relational database sold commercially by Oracle Corporation. Other databases, such as Informix™, DB2 (Database 2) or other data storage, including file-based, or query formats, platforms, or resources such as OLAP (On Line Analytical Processing), SQL (Structured Query Language), a SAN (storage area network), Microsoft Access™ or others may also be used, incorporated, or accessed. The database may comprise one or more such databases that reside in one or more physical devices and in one or more physical locations. The database may store a plurality of types of data and/or files and associated data or file descriptions, administrative information, or any other data.

Natural Language Processing

FIG. 2 depicts an NLP system 106 configured to perform natural language processing on an utterance, according to an implementation of the invention. NLP system 106 may include a an Automated Speech Recognition (“ASR”) engine 210, an Intent Recognition Engine (“IRE”) 220, text-to-speech engine 230, a dictionary 240, a user profile 250, an agent system 260, smart matching engine 270, confirmation engine 271, learning engine 272, execution engine 273, and/or other components. The components of NLP system 106 may communicate with one another via communication manager 202, which may mediate interactions between components of NLP system 106.

NLP system 106 may process an utterance encoded into audio to determine words or phrases from the utterance. Alternatively, NLP system 106 may process text that was already detected from the utterance. In whichever manner the utterance is received (e.g., whether as encoded audio or text), NLP system 106 may determine an intent of the utterance using the words or phrases, formulate requests based on its intent, and facilitate generating a response to or executing the request.

For example, in implementations in which an utterance is received as audio, ASR engine 210 may use words and phrases from dictionary 240 and information received from agent system 260 (discussed below) to recognize words from the utterance (i.e., the audio). The recognized words and phrases are processed by IRE 220, which determines the intent of the utterance and generates a request that corresponds to the utterance based on context, and information from agent system 260. The request may include a query (e.g., “what is the weather like in Seattle?”) a command (e.g., “turn up the volume”), and/or other information.

To process the intent of the utterance, IRE 220 may use keywords within the utterance itself to identify context. For example, for the utterance “what is the weather like in Seattle,” IRE 220 may determine that the context is weather and a subject is “Seattle.” In some instances, the contextual information may be based on immediately prior utterances. In these instances, IRE 220 may generate a stack that stores (e.g., in Random Access Memory) information relating to one or more previous utterances in a given session (which may be defined by utterances received within a threshold time of one another). For example, a first utterance “what is the weather like in Seattle” may be followed up with a second utterance “book me a flight there.” By itself, the second utterance “book me a flight there” may be non-specific because “there” is undefined in the second utterance. However, when processed as part of a stack that includes information relating to the first utterance “what is the weather like in Seattle,” IRE 220 determines that “there” refers to “Seattle.”

A natural language input may comprise a request and an entity name. A request may include several types of request, including a query, an executable command, and/or any other actionable statement. For example, a request may include a command to place a phone call or send a text message, may include a query for directions or a location, may include a question about the weather, and/or may include a question for a search engine. Numerous additional examples of requests may exist and may be processed from a received natural language input. Entity names may include objects of the processed request. For example, input entity names may include a contact to be called or texted, an address or establishment for which directions are sought, a location or time for which the weather is requested, and/or a target of a search engine query. Numerous other examples exist. Thus, a request and an input entity name, together may form a complete natural language input request, including both an action to be taken an object subject to that action. IRE 220 may process a received utterance to determine at least a type of request and an entity name.

Alternatively or additionally, NLP system 106 may use agent system 260, which includes system and domain specific agents, that assists in ASR, intent recognition, request generation, and/or request response. Agent system 260 may use nonvolatile storage (e.g., storage device 114 or storage device 144) to store data, parameters, history information, and locally stored content provided in the system databases 104 or other data sources. User specific data, parameters, and session and history information that may determine the behavior of the agents are stored in one or more user profiles 250. For example, user specific speech patterns may assist in ASR, while user preferences and history may assist in contextual determinations and, therefore, intent recognition.

The system agent provides default functionality and basic services. A domain specific agent relates to a corresponding domain of information (e.g., a business listing domain may relate to business listings, a contacts domain may relate to user contacts, a shopping domain may relate to commerce, etc.).

Agent system 260 create queries to local databases or though data sources on the Internet or other networks. Commands typically result in actions taken by the device on which NLP system 106 operates, or to a remote device or data source.

In implementations in which a response to a request is desired, agent system 260 may create a response string for presentation to the user. The response string may be presented as text and/or as speech, in which case the string is sent to the text to speech engine 230 to be output by a speaker (whether an on-board or off-board speaker).

Text-to-speech engine 230 may convert text to speech, such as when information from NLP system 106 (e.g., responses to spoken queries) is to be presented to a user in speech format. Conventional text-to-speech processors may be used for such text-to-speech generation.

A more detailed description of NLP processing is described in U.S. patent application Ser. No. 10/452,147, entitled “Systems and Methods for Responding Natural Language Speech Utterance,” filed on Jun. 3, 2003, the disclosure of which is incorporated in its entirety herein. NLP system 260 may use the foregoing and/or other types of NLP processing techniques and systems.

Smart matching engine 270 may include additional instructions that program computer system 110. The instructions of smart matching engine 270 may include, without limitation, instructions for suggesting named entities to appropriate portions of a received natural language input, and/or other instructions that program computer system 110 to perform various operations, each of which are described in greater detail herein.

Smart matching engine 270 may operate to identify an input entity name. ASR engine 210 and IRE 220 may process a natural language input to determine a type of request and an input entity name. Smart matching engine 270 may then match the input entity name to an entity name that can be acted upon. For example, if the user requests directions to Main St., then smart matching engine 270 may operate to suggest a match between the user uttered input entity name “Main St.” and an identified entity name—i.e., an actual, mappable, Main St. If the user asks to “call James,” then smart matching engine 270 may operate to suggest a match between an identified entity name, i.e., an actual contact in the user's directory, and the input entity name of “James.”

Smart matching engine 270 may suggest a match between an input entity name and at least one identified entity name based at least partially on an entity matching model. The entity matching model may comprise information compiled through machine learning techniques, and may include at least context information and user history information, as well as other information suitable for machine learning. A suggested match may include the input entity name and at least one identified entity name. In some embodiments, the at least one identified entity name may include a list of identified entity names. Thus, smart matching engine may identify multiple suggested matches between an input entity names A suggested match may further include additional entity match information.

Additional entity match information may include context information related to the natural language request, the input entity name, and the identified entity name. For example, for a user that lives in New York, an entity matching model may associate the entity name “taxi” in the request “call a taxi” with a particular taxi company in New York. When creating a suggested match for the input entity name “taxi,” for that user, smart matching engine 270 may include the additional entity match information of the user's location as part of the suggested match. This may permit the system to disambiguate the term “taxi” differently based on the user's location.

Context information usable by smart matching engine 270 may include a request context. For example, if a user makes a request to “call James” during the business day, smart matching engine 270 may suggest a match between the input entity name James and a business contact named James. If the request is made after hours, smart matching engine 270 may suggest a match with a contact from a friends and family contact list. Similarly, when attempting to disambiguate requests for directions, smart matching engine may suggest matches between locations that are closer to a present location of the user.

User history information usable by smart matching engine 270 may include information about previous user requests and activity. For example, when requesting to make a call, smart matching engine 270 may select a name from a user's contact list that the user calls very frequently. When seeking directions, smart matching engine 270 may suggest a destination location that matches a location the user has travelled to frequently.

Smart matching engine 270 may additionally employ machine learned matching techniques. For example, a user may frequently refer to a contact name by a nickname, making a request to “call Jim,” where the name “James” is stored in the user's contact list. Such a connection may become a part of an entity matching model based on machine learning techniques, as described in more detail below.

In some implementations smart matching engine 270 may suggest a match between an input entity name and an identified entity name based on an exact match. For example, where a user requests to “call John,” the system may suggest only matches that match “John” exactly. In some implementations smart matching engine 270 may suggest a match between an input entity name and an identified entity name based on a phonetic match. For example, where a user requests to “call John,” the system may suggest all matches that sound like “John,” including, for example, “Jon.” In some implementations smart matching engine 270 may suggest a match between an input entity name and an identified entity name based on similar sounds. Continuing to use the above example, smart matching engine 270 may suggest a match between “John” and “Juwan.” In some implementations smart matching engine 270 may suggest a match between an input entity name and an identified entity name based on an interpretation of a user's pronunciation style.

In some implementations smart matching engine 270 may suggest a match between an input entity name and an identified entity name based on learned associations. For example, a user may request to “call my brother.” Smart matching engine 270 may suggest a match between the input entity name “my brother,” and all user contacts having a same last name as the user. Smart matching engine 270 may also request clarification from the user about the entity name “my brother.” Once the system has successfully confirmed a match between a user contact and the input entity name “my brother,” the entity matching model may be updated with the information in order to correctly link the input entity name “my brother” to the appropriate contact. In another example, a user may request directions to “the pizza place.” Smart matching engine may propose a suggested match between the input entity name “the pizza place,” and the closest pizza restaurant. Once the system has successfully confirmed a match between “the pizza place,” and a particular pizza restaurant, the entity matching model may be updated to associate “the pizza place,” with the specific pizza restaurant.

In some implementations, smart matching engine 270 may pass a suggested match between an input named entity and an identified named entity to confirmation engine 271 for command confirmation, as described below.

In some implementations, smart matching engine 270 may determine an alternate identified entity name if confirmation engine 271 is unable to confirm the suggested match between the input named entity and the identified entity name. In some implementations, smart matching engine 270 may determine an alternate identified entity name using similar methods to those described above for selecting the first identified entity name. In determining an alternate identified entity name, smart matching engine 270 may select a next best match for an input entity name. In some implementations, smart matching engine 270 may develop an n-best list of suggested matches for an input entity name including n number of potential matches ranked in order of best fit. Smart matching engine 270 may provide the 2^nd, 3^rd, 4^thetc., matches from an n-best list, in order, to confirmation engine 271 as alternate identified names if the first suggestion is cancelled or met with non-confirmation. In some implementations, smart matching engine 270 may provide an entire n-best list to confirmation engine 271.

Confirmation engine 271 may include additional instructions that program computer system 110. The instructions of confirmation engine 271 may include, without limitation, instructions for confirming a suggest match between a received natural language input and a named entity, and/or other instructions that computer system 110 to perform various operations, each of which are described in greater detail herein.

Confirmation engine 271 may perform command confirmation based on a user's confirmation history of a proposed entity match. Command confirmation may be performed actively and/or passively. For example, active command confirmation may include querying the user to confirm a suggest match—e.g., where a user has asked to “call James,” a command confirmation may query the user “did you mean James Smith?” In some implementations, passive command confirmation may include permitting a user a predefined amount of time to cancel an action. For example, where a user has asked to “call James,” a command confirmation may alert the user “calling James Smith,” and wait for a predetermined amount of time to give the user a chance to cancel the action before executing the action. If the user does not cancel the action, then the suggested entity match may be confirmed. A predetermined amount of time may be relatively short, e.g., 1 to 5 seconds, or relatively long, e.g., 5-15 seconds. Selecting which command confirmation method to use may be based on a user's confirmation history.

For example, confirmation engine 271 may use a history of a user's confirmation between a particular input entity name and a particular identified entity name. For example, a user may ask to “call James,” and suggestion engine 270 may determine a suggested match between the input entity name “James” and an identified entity name “James Smith,” corresponding to a user contact. If suggestion engine 270 has previously suggested the same match, and the user has previously provided confirmations, confirmation engine 271 may alert the user of the impending action. If the user has not previously provided enough confirmations of the suggested match, confirmation engine 271 may request active confirmation from the user. Confirmation engine 271 may switch between these two methods when a number of previous confirmations of the suggested match exceeds a predetermined threshold number of times. Such a predetermined threshold number may be one or more times. A predetermined threshold number may vary depending on the type of entity name that is being matched (e.g., whether it is a contact name or a location address.)

In some implementations, confirmation engine 271 may update a number of previous confirmations of a suggested match in a user's confirmation history after confirmation or non-confirmation is received from the user. If confirmation is received from the user for a particular suggested match, confirmation engine 271 may increment the number of previous confirmations. If a user cancels or fails to confirm a suggested match, confirmation engine 271 may instead decrement the number of previous confirmations. In some implementations, confirmation engine 271 may reset a number of previous confirmations to zero after a non-confirmation is received. In some implementations, confirmation engine 271 may separately record a number of non-confirmations and decrement or reset the previous confirmation score after a threshold number of non-confirmations is reached. In some implementations, confirmation engine 271 may separately record a number of consecutive non-confirmations and decrement or reset the previous confirmation score after a threshold number of consecutive non-confirmations is reached. In some implementations, confirmation engine 271 may do nothing to a previous confirmation score when a non-confirmation is received.

In some implementations, confirmation engine 271 may request an alternate identified match from smart matching engine 270 if a non-confirmation or cancellation is received. Smart matching engine 270 may provide a next best alternate identified match to confirmation engine 271. Confirmation engine 271 may treat the alternate identified match similarly to the initial identified match, and determine an active or passive confirmation method based on a number of previous confirmations.

In some implementations, confirmation engine 271 may receive an n-best list from smart matching engine 270. Confirmation engine 271 may then try to confirm with the user each entry in the n-best list as a selected match. In some implementations, confirmation engine 271 may try each member of the n-best list in turn, using the above-discussed confirmation methods based on confirmation history. In some implementations, confirmation engine 271 may try to confirm the first entry of a an n-best list, and, if confirmation fails, query the user to select from the remaining entries in the n-best list.

Learning engine 272 may include additional instructions that program computer system 110. The instructions of learning engine 224 may include, without limitation, instructions for providing machine learning updates to a smart matching model used by smart matching engine 270 and confirmation engine 271 in suggesting and confirming matches between natural language inputs and named entities, and/or other instructions that computer system 110 to perform various operations, each of which are described in greater detail herein.

Learning engine 272 may include instructions for updating an entity matching model based on results provided by smart matching engine 270 and confirmation engine 271. As described above, smart matching engine 270 may suggest a match between an input entity name and an identified entity name, while confirmation engine 271 may operate to confirm the identified entity name and determine it as a selected entity name. When an entity name is confirmed as a selected entity name, learning engine 272 may update the entity matching model with information about the confirmed match. Learning engine 272 may update context information of the entity matching model with information about the context (e.g., time, location, whether the user is traveling, etc.) under which the suggested match was confirmed. Learning engine 272 may update user history information of the entity matching model, for example, to keep track of a frequency of user requests involving the selected entity name.

For example, where a user has a requested to make a phone call or send a text message to a contact and confirmation engine 271 confirms the suggested match provided by the smart matching engine 270, learning engine 272 may update the entity matching model with information about the match between the input entity name and the identified entity name, including, for example, time of day, location of the user, category of the entity name (e.g., business or personal contact), and any other suitable information. In some implementations, when confirmation engine 271 obtains a non-confirm result of a suggested match, learning engine 272 may update entity matching model with information that the match was not confirmed, as well as other context information associated with the failed match.

For example, a user may frequently his brother “James Lewis” in the evening but call his business associate “James Smith” during the day. Due to a frequency of calls, smart matching engine 270 may initially suggest a match between the request to “call James,” and “James Smith,” even when the call is taking place after 9 pm. After one or more non-confirmations between a command to “call James” at 9 pm and “James Smith,” learning engine 272 may sufficiently update the entity matching model to make an association between “James” and the user's brother “James Lewis” in the evening, and between “James” and the user's business associate “James Smith” during the day.

Execution engine 273 may include additional instructions that program computer system 110. The instructions of execution engine 273 may include, without limitation, instructions for executing a natural language request, and/or other instructions that computer system 110 to perform various operations, each of which are described in greater detail herein.

Execution engine 273 may be configured to cause a computer system on which it is instantiated to carry out the action indicated by the request and the selected entity name determined from the processed natural language input. Execution engine 273 may be configured to engage any systems of the device or computer system on which it is instantiated to carry out the natural language request.

Furthermore, as previously noted, some or all aspects of the foregoing operations may be performed by a given component illustrated in FIG. 1. In particular, some or all aspects of the foregoing operations may be performed by computer system 110 and/or user device 140. In some instances, some operations relating to natural language processing may be performed by user device 140, while other operations are performed at computer system 110.

In some instances, the same operation relating to natural language processing may be performed by both user device 140 and computer system 110 (i.e., both user device 140 and computer system 110 may perform one or more of the operations relating to natural language processing). In implementations where NLP system 106 operates in a hybrid manner (e.g., one or more components of NLP system 106 are included in and operate at different remote devices such as computer system 110 and user device 140), communication manager 202 may mediate communication through a network, such as network 102.

For example, a version of ASR engine 210 operating and executing on user device 140 may convert spoken utterances into text, which is then provided to a version of IRE 220 operating and executing on computer system 110. Alternatively or additionally, a version of ASR engine 210 operating and executing on user device 140 may convert spoken utterances into text and a version of ASR engine 210 operating and executing on computer system 110 may also convert the spoken utterances into text, in which case one result may be chosen over.

A more detailed description of such hybrid natural language processing is disclosed in U.S. patent application Ser. No. 12/703,032, entitled “System and Method for Hybrid Processing in a Natural Language Voice Services Environment,” filed on Feb. 9, 2010, the disclosure of which is incorporated in its entirety herein. The foregoing or other hybrid systems may be used for NLP system 106.

FIG. 3 depicts a process 300 of natural language processing disambiguation, according to an implementation of the invention. The various processing operations and/or data flows depicted in FIG. 3 (and in the other drawing figures) are described in greater detail herein. The described operations may be accomplished using some or all of the system components described in detail above and, in some implementations, various operations may be performed in different sequences and various operations may be omitted. Additional operations may be performed along with some or all of the operations shown in the depicted flow diagrams. One or more operations may be performed simultaneously. Accordingly, the operations as illustrated (and described in greater detail below) are exemplary by nature and, as such, should not be viewed as limiting.

In an operation 302, process 300 may include a natural language input operation. In natural language input operation 300 NLP system 106 receives a natural language input, such as a verbal utterance, from the user. The utterance may be received, for example, via a microphone incorporated into user device 140. ASR manager 210 of NLP subsystem 106 may receive the utterance.

In an operation 304, process 300 may include an natural language processing operation. Operation 304 may perform natural language processing on the received natural language input. Components of NLP subsystem 106 may cooperate to process the received natural language input to determine at least a request and an entity name. For example, as described above with respect to FIG. 2, ASR engine 210 may perform automatic speech recognition on the natural language input and IRE 220 may perform intent recognition on the natural language input. NLP subsystem 106 may thus determine a request and an input entity name from the user's natural language input.

In an operation 306, process 300 may include a smart matching step. During a smart matching operation 306, smart matching engine 270 may determine a suggested match between the input entity name and an identified entity name. Smart matching engine 270 may access an entity matching model, including context information, user history, and machine learned user tendencies to determine a suggested match.

In an operation 308, process 300 may include an entity match confirmation step. An entity match confirmation operation may be performed by confirmation engine 271. Confirmation engine 271 may determine whether to designate the identified entity name as a selected entity name, based on a user's confirmation history of suggested matches between the identified entity name and the input entity name. The user's confirmation history may be used to determine a method, e.g., active or passive, by which confirmation engine 271 confirms the entity name. Entity match confirmation operation 308 may be understood in more detail with reference to FIG. 4.

FIG. 4 depicts a process 400 of natural language match confirmation, according to an implementation of the invention. The various processing operations and/or data flows depicted in FIG. X (and in the other drawing figures) are described in greater detail herein. The described operations may be accomplished using some or all of the system components described in detail above and, in some implementations, various operations may be performed in different sequences and various operations may be omitted. Additional operations may be performed along with some or all of the operations shown in the depicted flow diagrams. One or more operations may be performed simultaneously. Accordingly, the operations as illustrated (and described in greater detail below) are exemplary by nature and, as such, should not be viewed as limiting.

In an operation 401, process 400 may include a suggested match reception step. During suggested match reception operation, confirmation engine 271 may receive information about a suggested match from smart matching operation 270. Confirmation engine 271 may receive the input entity name and the suggested match for it, i.e., the identified entity name.

In an operation 402, process 400 may include a match confirmation history determination step. Match confirmation history determination operation 402 may access a user history database 180/190 to determine a user command confirmation history associated with the suggested match. The user command confirmation history may include a previous confirmations counter comprising information about a user's previous confirmations of the suggested match between the input entity name and identified entity name. For example, a user's confirmation history may include the information that a user has confirmed a particular match 9 times, cancelled that match 2 times, and confirmed the previous 6 suggestions of the match.

In some implementations, the user command confirmation history may further include additional entity match information, which may include the context of a user's confirmation history with respect to a suggested match. For example, additional entity match information may include information about circumstances under which a user has confirmed a suggested match. For example, where a user has made the request to “call James,” and the suggested match is with the identified entity “James Smith” the business associate, the additional entity match information may include the information that the user has confirmed this suggested match 13 times during business hours, but cancelled the match 7 times during evening hours. During match confirmation history determination operation 402, confirmation engine 271 may determine the number of previous confirmations by a user of the suggested match between the input entity name and the identified entity name, and may further use additional entity match information to make the determination.

In an operation 404, process 400 may include a confirmation history thresholding step. Confirmation history thresholding operation 404 may include comparing the number of previous confirmations by the user to a predetermined threshold number of confirmations. Confirmation history thresholding operation 404 may include determining whether the previous number of confirmations exceeds or fails to exceed the predetermined threshold number of confirmations.

In an operation 406, process 400 may include an active match confirmation step. If confirmation engine 271 determines that a number of previous confirmations does not exceed a predetermined confirmation threshold number, process 400 may proceed to an active confirmation request operation 406. During active confirmation request operation 406, confirmation engine 271 may cause the host system to query the use for active confirmation of the suggested match provided by smart matching engine 270. For example, where a user has requested directions to the input entity name “233 Main St.,” the system may actively request that the user confirm the identified entity name by querying the user “Did you mean 233 Main St, Cambridge, Mass.?” When a match has been confirmed by the user, confirmation engine 271 may designate the confirmed entity name as a selected entity name for subsequent execution. Confirmation engine 271 may further update the previous confirmation history with the information that the suggested match has been confirmed. Such updating may include incrementing a number of previous confirmations, incrementing a number of previous consecutive confirmations, and/or providing any additional entity match information about the context in which the suggested match was confirmed.

When a match has been cancelled or decline by the user, confirmation engine 271 may designate the identified entity name as a non-match, and proceed to an alternative match suggestion operation 410. Confirmation engine 271 may further update the previous confirmation history with the information that the suggested match was not confirmed. Such updating may include decrementing a number of previous confirmations, resetting a number of previous consecutive confirmations, resetting a total number of previous confirmations, and/or providing any additional entity match information about the context in which the suggested match was cancelled.

In an operation 410, process 400 may include an alternative match suggestion step. Confirmation engine 271 may perform an alternative match suggestion operation after a user has cancelled or failed to confirm an initially suggested match provided by smart matching engine 270 (either in operation 406 or an operation 412). In some implementations, confirmation engine 271 may request that smart matching engine 270 provide a next best match for the input entity name. Confirmation engine 271 may then suggest this next best match as an alternate identified entity name to the user for confirmation. In some implementations, confirmation engine 271 may select a next best match from an n-best list that has already been provided by smart matching engine 270. Confirmation engine 271 may then suggest this next best match as an alternate identified entity name to the user for confirmation. In some implementations, confirmation engine 271 may provide, to the user, multiple potential matches, e.g., from an n-best list provided by smart matching engine 270, from which the user may select an alternate identified entity name for confirmation. After alternate match suggestion step 410, process flow may return to catch confirmation history determination operation 402 and repeat steps 402, 404, 406, and/or 412 using the alternate identified entity name.

In an operation 412, process 400 may include a passive confirmation operation. If confirmation engine 271, at thresholding step 404, determines that a number of previous confirmations exceeds a predetermined confirmation threshold number, process 400 may proceed to an passive confirmation request operation 412. A passive confirmation operation 412 may be performed by confirmation engine 271 when the user has previously confirmed the suggested match more than a certain number of times. After multiple past confirmations, the system may determine a higher level of certainty in the suggested match, and not require that the user actively confirm the match. A passive confirmation operation may proceed by alerting the user (e.g., by an audio prompt or a display prompt) that an action based on the request and the identified entity name will be carried out. A predetermined pause period may then be introduced to give the opportunity to the user to cancel the action. If the user does not cancel the action, then confirmation engine 271 may confirm the suggested match. If the user does issue a cancellation command, then confirmation engine 271 may designate the suggested match as non-confirmed.

When a match has been confirmed by the user, confirmation engine 271 may designate the confirmed entity name as a selected entity name for subsequent execution at operation 414. Confirmation engine 271 may further update the previous confirmation history with the information that the suggested match has been confirmed. Such updating may include incrementing a number of previous confirmations, incrementing a number of previous consecutive confirmations, and/or providing any additional entity match information about the context in which the suggested match was confirmed.

When a match has been cancelled or non-confirmed by the user, confirmation engine 271 may designate the identified entity name as a non-match, and proceed to an alternative match suggestion operation 410. Confirmation engine 271 may further update the previous confirmation history with the information that the suggested match was not confirmed. Such updating may include decrementing a number of previous confirmations, resetting a number of previous consecutive confirmations, resetting a total number of previous confirmations, and/or providing any additional entity match information about the context in which the suggested match was cancelled.

In an operation 414, process 400 may include a selected match finalization operation. Selected match finalization operation 414 may be performed by confirmation engine 271 to pass the confirmed selected entity name to execution engine 273. During operation 414, information about the confirmed match, e.g., the selected entity name and any additional match entity information, may be passed to learning engine 272 to update the entity matching model.

Returning now to FIG. 3, in an operation 310, process 310 may include a natural language execution step. After confirmation of the identified entity name as the selected entity name during command confirmation operation 308, natural language execution operation 310 may be performed by execution engine 273 to cause the system to carry out the request on the selected entity name. For example, the request may include a phone dialing request and the selected entity name may be a user object. Execution of this natural language result would cause a device of the user to initiate a phone call. The request may include a search request, and the selected entity name may include a search object. Execution of this natural language result would cause a device of the user to carry out a search for information about the search object. The request may include a navigation request, and the selected entity name may include a street address or other physical location. Execution of this statement would cause the device of the user to provide the user with directions to the street address or physical location.

Other implementations, uses and advantages of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. The specification should be considered exemplary only, and the scope of the invention is accordingly intended to be limited only by the following claims.

Claims

1. A computer implemented method of natural language disambiguation, the method being implemented in a computer system having one or more physical processors programmed with computer program instructions that, when executed by the one or more physical processors, cause the computer system to perform the method, the method comprising: processing, by the computer system, using an intent recognition engine, a natural language input;determining, by the computer system, based on the processing, at least a type of request and an input entity name;identifying, by the computer system, at least a first match between the input entity name and a first identified entity name and at least a second match between the input entity name and a second identified entity name;selectively performing, by the computer system, one of a passive confirmation or an active confirmation of whether the first identified entity name or the second identified entity name was intended by the natural language input, depending on a threshold comparison applied to previous confirmation history of a number of previous confirmations of whether the first identified entity name or the second entity name was intended by the natural language input, wherein the active confirmation comprises: requesting, by the computer system, a confirmation of whether the first identified entity name or the second identified entity name was intended by the natural language input; andreceiving, by the computer system, a response to the requested confirmation;wherein the passive confirmation comprises: selecting, as a response, the first identified entity name or the second identified entity name as the intended entity name;generating, by the computer system, a request based on the type of request and the response; andupdating the number of previous responses regarding whether the first identified entity name or the second identified entity name was intended by the natural language input.
2. The computer implemented method of claim 1, wherein the input entity name includes at least one of a contact name, an address, or a business name.
3. The computer implemented method of claim 1, wherein the type of request includes a phone dialing request and the first identified entity name or the second identified entity name includes a user contact.
4. The computer implemented method of claim 1, wherein the type of request includes a search query and the first identified entity name or the second identified entity name includes a search object.
5. The computer implemented method of claim 1, wherein the type of request includes a navigation query and the first identified entity name or the second identified entity name includes a street address.
6. The computer implemented method of claim 1, wherein identifying at least a first match between the input entity name and the first identified entity name and at least a second match between the input entity name and the second identified entity name is at least partially based on at least one of a user location, a time of day, a user request history, a user request frequency, and a user preference.
7. The computer implemented method of claim 1, wherein identifying at least a first match between the input entity name and the first identified entity name and at least a second match between the input entity name and the second identified entity name is at least partially based on a user preference for contacting at least one of family, friends, and business associates.
8. The computer implemented method of claim 1, wherein: the active confirmation is performed if a number of previous confirmations of whether the first identified entity name or the second identified entity name was intended by the natural language input is below a predetermined threshold; andwherein the passive confirmation is performed if a number of previous confirmations of whether the first identified entity name or the second identified entity name was intended by the natural language input is above a predetermined threshold.
9. The computer implemented method of claim 1, wherein the method further comprises:alerting the user that execution of the request will be performed;waiting, by the computer system, a predetermined amount of time for user cancellation of the announced execution prior to executing the request; andupdating the number of previous responses regarding whether the first identified entity name or the second identified entity name was intended by the natural language input.
10. The computer implemented method of claim 9, further comprising: receiving, by the computer system, a cancellation command from the user;requesting that the user select an alternate request different from the generated request; andstoring the alternate request selected by the user for use in future natural language input processing.
11. The computer implemented method of claim 1, wherein the method further comprises:processing, by the computer system, using the intent recognition engine, a second natural language input;determining, by the computer system, a second input entity name based on the processing;identifying, by the computer system, at least a third match between the second input entity name and the first identified entity name and at least a fourth match between the second input entity name and the second identified entity name;determining, by the computer system, that the first number of times that the first identified entity name was intended exceeds a predetermined threshold; anddetermining, by the computer system, that the first identified entity name is intended by the second natural language input based on the determination that the first number of times that the first identified entity name was intended exceeds the predetermined threshold.
12. The computer implemented method of claim 1, wherein requesting a confirmation comprises: requesting that the user provide an alternate entity name, different from the first identified entity name and the second identified entity name;providing information about the alternate entity name as the response to the requested confirmation,and wherein the method further comprises:storing, by the computer system, information indicating the response for use in future natural language input processing.
13. The computer implemented method of claim 1, further comprising: updating the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being an exact match.
14. The computer implemented method of claim 1, further comprising: updating the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being a phonetic match.
15. The computer implemented method of claim 1, further comprising: updating the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being a partial match.
16. A system of natural language disambiguation, the system comprising: a computer system comprising one or more physical processors programmed by computer program instructions that, when executed, cause the computer system to:process, using an intent recognition engine, a natural language input;determine, based on the processed natural language input, at least a type of request and an input entity name;identify at least a first match between the input entity name and a first identified entity name and at least a second match between the input entity name and a second identified entity name;selectively perform one of a passive confirmation or an active confirmation of whether the first identified entity name or the second identified entity name was intended by the natural language input, depending on a threshold comparison applied to previous confirmation history of a number of previous confirmations of whether the first identified entity name or the second entity name was intended by the natural language input, wherein the active confirmation comprises: requesting a confirmation of whether the first identified entity name or the second identified entity name was intended by the natural language input; andreceiving a response to the requested confirmation;wherein the passive confirmation comprises: selecting, as a response, the first identified entity name or the second identified entity name as the intended entity name;generate a request based on the type of request and the response; andupdate the number of previous responses regarding whether the first identified entity name or the second identified entity name was intended by the natural language input.
17. The system of claim 16, wherein the input entity name includes at least one of a contact name, an address, and a business name.
18. The system of claim 16, wherein the request includes a phone dialing request and the first identified entity name or the second identified entity name includes a user contact.
19. The system of claim 16, wherein the type of request includes a search query and the first identified entity name or the second identified entity name includes a search object.
20. The system of claim 16, wherein the type of request includes a navigation query and the first identified entity name or the second identified entity name includes a street address.
21. The system of claim 16, wherein the computer program instructions further cause the computer system to: identify at least the first match between the input entity name and the first identified entity name and at least the second match between the input entity name and the second identified entity name at least partially based on at least one of a user location, a time of day, a user request history, a user request frequency, and a user preference.
22. The system of claim 16, wherein the computer program instructions further cause the computer system to: identify at least the first match between the input entity name and the first identified entity name and at least the second match between the input entity name and the second identified entity name at least partially based on user preferences including a preference for contacting at least one of family, friends, and business associates.
23. The system of claim 16, wherein the computer program instructions further cause the computer system to: perform the active confirmation if a number of previous confirmations of whether the first identified entity name or the second identified entity name was intended by the natural language input is below a predetermined threshold; andperform the passive confirmation if a number of previous confirmations of whether the first identified entity name or the second identified entity name was intended by the natural language input is above a predetermined threshold.
24. The system of claim 16, wherein the computer program instructions further cause the computer system to:alert the user that execution of the intended request will be performed;wait a predetermined amount of time for user cancellation of the announced execution prior to executing the intended request; andupdate the number of previous responses regarding whether the first identified entity name or the second identified entity name was intended by the natural language input.
25. The system of claim 24, further comprising instructions to cause the computer system to: receive a cancellation command from the user;request that the user select an alternate request different from the generated request; andstore the alternate request selected by the user for use in future natural language input processing.
26. The system of claim 16, wherein the computer program instructions further cause the computer system to:process, using the intent recognition engine, a second natural language input;identify a second input entity name based on the processing;identify at least a third match between the second input entity name and the first identified entity name and at least a fourth match between the second input entity name and the second identified entity name;determine that the first number of times that the first identified entity name was intended exceeds a predetermined threshold; anddetermine that the first identified entity name is intended by the second natural language input based on the determination that the first number of times that the first identified entity name was intended exceeds the predetermined threshold.
27. The system of claim 16, wherein the computer program instructions to request a confirmation further cause the computer system to: request that the user provide an alternate entity name, different from the first identified entity name and the second identified entity name;provide information about the alternate entity name as the response to the requested confirmation, andwherein the computer system is further caused to store information indicating the response for use in future natural language input processing.
28. The system of claim 16, wherein the computer program instructions further cause the computer system to: update the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being an exact match.
29. The system of claim 16, wherein the computer program instructions further cause the computer system to: update the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being a phonetic match.
30. The system of claim 16, wherein the computer program instructions further cause the computer system to: update the intent recognition engine with information indicating a match between the input entity name and the first identified entity name or second identified entity name based on one of the passive confirmation or the active confirmation,wherein the computer system recognizes the first identified entity name and the second identified entity name as being a partial match.

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Patent Application Ser. No. 62/368,975, entitled “SYSTEM AND METHOD OF DISAMBIGUATING NATURAL LANGUAGE PROCESSING REQUESTS”, filed Jul. 29, 2016, which is hereby incorporated herein by reference in its entirety.

US Referenced Citations (902)

Number	Name	Date	Kind
4430669	Cheung	Feb 1984	A
4821027	Mallory	Apr 1989	A
4829423	Tennant	May 1989	A
4887212	Zamora	Dec 1989	A
4910784	Doddington	Mar 1990	A
5027406	Roberts	Jun 1991	A
5155743	Jacobs	Oct 1992	A
5164904	Sumner	Nov 1992	A
5208748	Flores	May 1993	A
5265065	Turtle	Nov 1993	A
5274560	LaRue	Dec 1993	A
5331554	Graham	Jul 1994	A
5357596	Takebayashi	Oct 1994	A
5369575	Lamberti	Nov 1994	A
5377350	Skinner	Dec 1994	A
5386556	Hedin	Jan 1995	A
5424947	Nagao	Jun 1995	A
5471318	Ahuja	Nov 1995	A
5475733	Eisdorfer	Dec 1995	A
5479563	Yamaguchi	Dec 1995	A
5488652	Bielby	Jan 1996	A
5499289	Bruno	Mar 1996	A
5500920	Kupiec	Mar 1996	A
5517560	Greenspan	May 1996	A
5533108	Harris	Jul 1996	A
5537436	Bottoms	Jul 1996	A
5539744	Chu	Jul 1996	A
5557667	Bruno	Sep 1996	A
5559864	Kennedy, Jr.	Sep 1996	A
5563937	Bruno	Oct 1996	A
5577165	Takebayashi	Nov 1996	A
5590039	Ikeda	Dec 1996	A
5608635	Tamai	Mar 1997	A
5615296	Stanford	Mar 1997	A
5617407	Bareis	Apr 1997	A
5633922	August	May 1997	A
5634086	Rtischev	May 1997	A
5652570	Lepkofker	Jul 1997	A
5675629	Raffel	Oct 1997	A
5696965	Dedrick	Dec 1997	A
5708422	Blonder	Jan 1998	A
5721938	Stuckey	Feb 1998	A
5722084	Chakrin	Feb 1998	A
5740256	CastelloDaCosta	Apr 1998	A
5742763	Jones	Apr 1998	A
5748841	Morin	May 1998	A
5748974	Johnson	May 1998	A
5752052	Richardson	May 1998	A
5754784	Garland	May 1998	A
5761631	Nasukawa	Jun 1998	A
5774841	Salazar	Jun 1998	A
5774859	Houser	Jun 1998	A
5794050	Dahlgren	Aug 1998	A
5794196	Yegnanarayanan	Aug 1998	A
5797112	Komatsu	Aug 1998	A
5799276	Komissarchik	Aug 1998	A
5802510	Jones	Sep 1998	A
5829000	Huang	Oct 1998	A
5832221	Jones	Nov 1998	A
5839107	Gupta	Nov 1998	A
5848396	Gerace	Dec 1998	A
5855000	Waibel	Dec 1998	A
5867817	Catallo	Feb 1999	A
5878385	Bralich	Mar 1999	A
5878386	Coughlin	Mar 1999	A
5892813	Morin	Apr 1999	A
5892900	Ginter	Apr 1999	A
5895464	Bhandari	Apr 1999	A
5895466	Goldberg	Apr 1999	A
5897613	Chan	Apr 1999	A
5899991	Karch	May 1999	A
5902347	Backman	May 1999	A
5911120	Jarett	Jun 1999	A
5918222	Fukui	Jun 1999	A
5926784	Richardson	Jul 1999	A
5933822	Braden-Harder	Aug 1999	A
5950167	Yaker	Sep 1999	A
5953393	Culbreth	Sep 1999	A
5960384	Brash	Sep 1999	A
5960397	Rahim	Sep 1999	A
5960399	Barclay	Sep 1999	A
5960447	Holt	Sep 1999	A
5963894	Richardson	Oct 1999	A
5963940	Liddy	Oct 1999	A
5983190	Trower, II	Nov 1999	A
5987404	DellaPietra	Nov 1999	A
5991721	Asano	Nov 1999	A
5995119	Cosatto	Nov 1999	A
5995928	Nguyen	Nov 1999	A
5995943	Bull	Nov 1999	A
6009382	Martino	Dec 1999	A
6014559	Amin	Jan 2000	A
6018708	Dahan	Jan 2000	A
6021384	Gorin	Feb 2000	A
6028514	Lemelson	Feb 2000	A
6035267	Watanabe	Mar 2000	A
6044347	Abella	Mar 2000	A
6049602	Foladare	Apr 2000	A
6049607	Marash	Apr 2000	A
6058187	Chen	May 2000	A
6067513	Ishimitsu	May 2000	A
6073098	Buchsbaum	Jun 2000	A
6076059	Glickman	Jun 2000	A
6078886	Dragosh	Jun 2000	A
6081774	deHita	Jun 2000	A
6085186	Christianson	Jul 2000	A
6101241	Boyce	Aug 2000	A
6108631	Ruhl	Aug 2000	A
6119087	Kuhn	Sep 2000	A
6119101	Peckover	Sep 2000	A
6122613	Baker	Sep 2000	A
6134235	Goldman	Oct 2000	A
6144667	Doshi	Nov 2000	A
6144938	Surace	Nov 2000	A
6154526	Dahlke	Nov 2000	A
6160883	Jackson	Dec 2000	A
6167377	Gillick	Dec 2000	A
6173266	Marx	Jan 2001	B1
6173279	Levin	Jan 2001	B1
6175858	Bulfer	Jan 2001	B1
6185535	Hedin	Feb 2001	B1
6188982	Chiang	Feb 2001	B1
6192110	Abella	Feb 2001	B1
6192338	Haszto	Feb 2001	B1
6195634	Dudemaine	Feb 2001	B1
6195651	Handel	Feb 2001	B1
6199043	Happ	Mar 2001	B1
6208964	Sabourin	Mar 2001	B1
6208972	Grant	Mar 2001	B1
6219346	Maxemchuk	Apr 2001	B1
6219643	Cohen	Apr 2001	B1
6219645	Byers	Apr 2001	B1
6226612	Srenger	May 2001	B1
6233556	Teunen	May 2001	B1
6233559	Balakrishnan	May 2001	B1
6233561	Junqua	May 2001	B1
6236968	Kanevsky	May 2001	B1
6243679	Mohri	Jun 2001	B1
6246981	Papineni	Jun 2001	B1
6246990	Happ	Jun 2001	B1
6266636	Kosaka	Jul 2001	B1
6269336	Ladd	Jul 2001	B1
6272455	Hoshen	Aug 2001	B1
6272461	Meredith	Aug 2001	B1
6275231	Obradovich	Aug 2001	B1
6278377	DeLine	Aug 2001	B1
6278968	Franz	Aug 2001	B1
6286002	Axaopoulos	Sep 2001	B1
6288319	Catona	Sep 2001	B1
6292767	Jackson	Sep 2001	B1
6301560	Masters	Oct 2001	B1
6308151	Smith	Oct 2001	B1
6311159	VanTichelen	Oct 2001	B1
6314402	Monaco	Nov 2001	B1
6321196	Franceschi	Nov 2001	B1
6356869	Chapados	Mar 2002	B1
6362748	Huang	Mar 2002	B1
6366882	Bijl	Apr 2002	B1
6366886	Dragosh	Apr 2002	B1
6374214	Friedland	Apr 2002	B1
6374226	Hunt	Apr 2002	B1
6377913	Coffman	Apr 2002	B1
6381535	Durocher	Apr 2002	B1
6385596	Wiser	May 2002	B1
6385646	Brown	May 2002	B1
6389398	Lustgarten	May 2002	B1
6393403	Majaniemi	May 2002	B1
6393428	Miller	May 2002	B1
6397181	Li	May 2002	B1
6404878	Jackson	Jun 2002	B1
6405170	Phillips	Jun 2002	B1
6408272	White	Jun 2002	B1
6411810	Maxemchuk	Jun 2002	B1
6411893	Ruhl	Jun 2002	B2
6415257	Junqua	Jul 2002	B1
6418210	Sayko	Jul 2002	B1
6420975	DeLine	Jul 2002	B1
6429813	Feigen	Aug 2002	B2
6430285	Bauer	Aug 2002	B1
6430531	Polish	Aug 2002	B1
6434523	Monaco	Aug 2002	B1
6434524	Weber	Aug 2002	B1
6434529	Walker	Aug 2002	B1
6442522	Carberry	Aug 2002	B1
6446114	Bulfer	Sep 2002	B1
6453153	Bowker	Sep 2002	B1
6453292	Ramaswamy	Sep 2002	B2
6456711	Cheung	Sep 2002	B1
6456974	Baker	Sep 2002	B1
6466654	Cooper	Oct 2002	B1
6466899	Yano	Oct 2002	B1
6470315	Netsch	Oct 2002	B1
6487494	Odinak	Nov 2002	B2
6487495	Gale	Nov 2002	B1
6498797	Anerousis	Dec 2002	B1
6499013	Weber	Dec 2002	B1
6501833	Phillips	Dec 2002	B2
6501834	Milewski	Dec 2002	B1
6505155	Vanbuskirk	Jan 2003	B1
6510417	Woods	Jan 2003	B1
6513006	Howard	Jan 2003	B2
6522746	Marchok	Feb 2003	B1
6523061	Halverson	Feb 2003	B1
6532444	Weber	Mar 2003	B1
6539348	Bond	Mar 2003	B1
6549629	Finn	Apr 2003	B2
6553372	Brassell	Apr 2003	B1
6556970	Sasaki	Apr 2003	B1
6556973	Lewin	Apr 2003	B1
6560576	Cohen	May 2003	B1
6560590	Shwe	May 2003	B1
6567778	ChaoChang	May 2003	B1
6567797	Schuetze	May 2003	B1
6567805	Johnson	May 2003	B1
6570555	Prevost	May 2003	B1
6570964	Murveit	May 2003	B1
6571279	Herz	May 2003	B1
6574597	Mohri	Jun 2003	B1
6574624	Johnson	Jun 2003	B1
6578022	Foulger	Jun 2003	B1
6581103	Dengler	Jun 2003	B1
6584439	Geilhufe	Jun 2003	B1
6587858	Strazza	Jul 2003	B1
6591185	Polidi	Jul 2003	B1
6591239	McCall	Jul 2003	B1
6594257	Doshi	Jul 2003	B1
6594367	Marash	Jul 2003	B1
6598018	Junqua	Jul 2003	B1
6601026	Appelt	Jul 2003	B2
6601029	Pickering	Jul 2003	B1
6604075	Brown	Aug 2003	B1
6604077	Dragosh	Aug 2003	B2
6606598	Holthouse	Aug 2003	B1
6611692	Raffel	Aug 2003	B2
6614773	Maxemchuk	Sep 2003	B1
6615172	Bennett	Sep 2003	B1
6622119	Ramaswamy	Sep 2003	B1
6629066	Jackson	Sep 2003	B1
6631346	Karaorman	Oct 2003	B1
6631351	Ramachandran	Oct 2003	B1
6633846	Bennett	Oct 2003	B1
6636790	Lightner	Oct 2003	B1
6643620	Contolini	Nov 2003	B1
6647363	Claassen	Nov 2003	B2
6650747	Bala	Nov 2003	B1
6658388	Kleindienst	Dec 2003	B1
6678680	Woo	Jan 2004	B1
6681206	Gorin	Jan 2004	B1
6691151	Cheyer	Feb 2004	B1
6701294	Ball	Mar 2004	B1
6704396	Parolkar	Mar 2004	B2
6704576	Brachman	Mar 2004	B1
6704708	Pickering	Mar 2004	B1
6707421	Drury	Mar 2004	B1
6708150	Hirayama	Mar 2004	B1
6721001	Berstis	Apr 2004	B1
6721633	Funk	Apr 2004	B2
6721706	Strubbe	Apr 2004	B1
6726636	DerGhazarian	Apr 2004	B2
6732088	Glance	May 2004	B1
6735592	Neumann	May 2004	B1
6739556	Langston	May 2004	B1
6741931	Kohut	May 2004	B1
6742021	Halverson	May 2004	B1
6745161	Arnold	Jun 2004	B1
6751591	Gorin	Jun 2004	B1
6751612	Schuetze	Jun 2004	B1
6754485	Obradovich	Jun 2004	B1
6754627	Woodward	Jun 2004	B2
6754647	Tackett	Jun 2004	B1
6757544	Rangarajan	Jun 2004	B2
6757718	Halverson	Jun 2004	B1
6785651	Wang	Aug 2004	B1
6795808	Strubbe	Sep 2004	B1
6801604	Maes	Oct 2004	B2
6801893	Backfried	Oct 2004	B1
6804330	Jones	Oct 2004	B1
6810375	Ejerhed	Oct 2004	B1
6813341	Mahoney	Nov 2004	B1
6816830	Kempe	Nov 2004	B1
6829603	Chai	Dec 2004	B1
6832230	Zilliacus	Dec 2004	B1
6833848	Wolff	Dec 2004	B1
6850603	Eberle	Feb 2005	B1
6856990	Barile	Feb 2005	B2
6865481	Kawazoe	Mar 2005	B2
6868380	Kroeker	Mar 2005	B2
6868385	Gerson	Mar 2005	B1
6871179	Kist	Mar 2005	B1
6873837	Yoshioka	Mar 2005	B1
6877001	Wolf	Apr 2005	B2
6877134	Fuller	Apr 2005	B1
6882970	Garner	Apr 2005	B1
6901366	Kuhn	May 2005	B1
6910003	Arnold	Jun 2005	B1
6912498	Stevens	Jun 2005	B2
6915126	Mazzara, Jr.	Jul 2005	B2
6928614	Everhart	Aug 2005	B1
6934756	Maes	Aug 2005	B2
6937977	Gerson	Aug 2005	B2
6937982	Kitaoka	Aug 2005	B2
6941266	Gorin	Sep 2005	B1
6944594	Busayapongchai	Sep 2005	B2
6950821	Faybishenko	Sep 2005	B2
6954755	Reisman	Oct 2005	B2
6959276	Droppo	Oct 2005	B2
6961700	Mitchell	Nov 2005	B2
6963759	Gerson	Nov 2005	B1
6964023	Maes	Nov 2005	B2
6968311	Knockeart	Nov 2005	B2
6973387	Masclet	Dec 2005	B2
6975993	Keiller	Dec 2005	B1
6980092	Turnbull	Dec 2005	B2
6983055	Luo	Jan 2006	B2
6990513	Belfiore	Jan 2006	B2
6996531	Korall	Feb 2006	B2
7003463	Maes	Feb 2006	B1
7016849	Arnold	Mar 2006	B2
7020609	Thrift	Mar 2006	B2
7024364	Guerra	Apr 2006	B2
7027586	Bushey	Apr 2006	B2
7027974	Busch	Apr 2006	B1
7027975	Pazandak	Apr 2006	B1
7035415	Belt	Apr 2006	B2
7036128	Julia	Apr 2006	B1
7043425	Pao	May 2006	B2
7054817	Shao	May 2006	B2
7058890	George	Jun 2006	B2
7062488	Reisman	Jun 2006	B1
7069220	Coffman	Jun 2006	B2
7072834	Zhou	Jul 2006	B2
7072888	Perkins	Jul 2006	B1
7076362	Ohtsuji	Jul 2006	B2
7082469	Gold	Jul 2006	B2
7085708	Manson	Aug 2006	B2
7092928	Elad	Aug 2006	B1
7107210	Deng	Sep 2006	B2
7107218	Preston	Sep 2006	B1
7110951	Lemelson	Sep 2006	B1
7127395	Gorin	Oct 2006	B1
7127400	Koch	Oct 2006	B2
7130390	Abburi	Oct 2006	B2
7136875	Anderson	Nov 2006	B2
7137126	Coffman	Nov 2006	B1
7143037	Chestnut	Nov 2006	B1
7143039	Stifelman	Nov 2006	B1
7146319	Hunt	Dec 2006	B2
7149696	Shimizu	Dec 2006	B2
7165028	Gong	Jan 2007	B2
7170993	Anderson	Jan 2007	B2
7171291	Obradovich	Jan 2007	B2
7174300	Bush	Feb 2007	B2
7177798	Hsu	Feb 2007	B2
7184957	Brookes	Feb 2007	B2
7190770	Ando	Mar 2007	B2
7197069	Agazzi	Mar 2007	B2
7197460	Gupta	Mar 2007	B1
7203644	Anderson	Apr 2007	B2
7206418	Yang	Apr 2007	B2
7207011	Mulvey	Apr 2007	B2
7215941	Beckmann	May 2007	B2
7228276	Omote	Jun 2007	B2
7231343	Treadgold	Jun 2007	B1
7236923	Gupta	Jun 2007	B1
7254482	Kawasaki	Aug 2007	B2
7272212	Eberle	Sep 2007	B2
7277854	Bennett	Oct 2007	B2
7283829	Christenson	Oct 2007	B2
7283951	Marchisio	Oct 2007	B2
7289606	Sibal	Oct 2007	B2
7299186	Kuzunuki	Nov 2007	B2
7301093	Sater	Nov 2007	B2
7305381	Poppink	Dec 2007	B1
7321850	Wakita	Jan 2008	B2
7328155	Endo	Feb 2008	B2
7337116	Charlesworth	Feb 2008	B2
7340040	Saylor	Mar 2008	B1
7366285	Parolkar	Apr 2008	B2
7366669	Nishitani	Apr 2008	B2
7376645	Bernard	May 2008	B2
7386443	Parthasarathy	Jun 2008	B1
7398209	Kennewick	Jul 2008	B2
7406421	Odinak	Jul 2008	B2
7415100	Cooper	Aug 2008	B2
7415414	Azara	Aug 2008	B2
7421393	DiFabbrizio	Sep 2008	B1
7424431	Greene	Sep 2008	B2
7447635	Konopka	Nov 2008	B1
7451088	Ehlen	Nov 2008	B1
7454368	Stillman	Nov 2008	B2
7454608	Gopalakrishnan	Nov 2008	B2
7461059	Richardson	Dec 2008	B2
7472020	Brulle-Drews	Dec 2008	B2
7472060	Gorin	Dec 2008	B1
7472075	Odinak	Dec 2008	B2
7477909	Roth	Jan 2009	B2
7478036	Shen	Jan 2009	B2
7487088	Gorin	Feb 2009	B1
7487110	Bennett	Feb 2009	B2
7493259	Jones	Feb 2009	B2
7493559	Wolff	Feb 2009	B1
7502672	Kolls	Mar 2009	B1
7502730	Wang	Mar 2009	B2
7502738	Kennewick	Mar 2009	B2
7516076	Walker	Apr 2009	B2
7529675	Maes	May 2009	B2
7536297	Byrd	May 2009	B2
7536374	Au	May 2009	B2
7542894	Murata	Jun 2009	B2
7546382	Healey	Jun 2009	B2
7548491	Macfarlane	Jun 2009	B2
7552054	Stifelman	Jun 2009	B1
7558730	Davis	Jul 2009	B2
7574362	Walker	Aug 2009	B2
7577244	Taschereau	Aug 2009	B2
7606708	Hwang	Oct 2009	B2
7606712	Smith	Oct 2009	B1
7620549	DiCristo	Nov 2009	B2
7634409	Kennewick	Dec 2009	B2
7640006	Portman	Dec 2009	B2
7640160	Di Cristo	Dec 2009	B2
7640272	Mahajan	Dec 2009	B2
7672931	Hurst-Hiller	Mar 2010	B2
7676365	Hwang	Mar 2010	B2
7676369	Fujimoto	Mar 2010	B2
7684977	Morikawa	Mar 2010	B2
7693720	Kennewick	Apr 2010	B2
7697673	Chiu	Apr 2010	B2
7706616	Kristensson	Apr 2010	B2
7729916	Coffman	Jun 2010	B2
7729918	Walker	Jun 2010	B2
7729920	Chaar	Jun 2010	B2
7734287	Ying	Jun 2010	B2
7748021	Obradovich	Jun 2010	B2
7788084	Brun	Aug 2010	B2
7792257	Vanier	Sep 2010	B1
7801731	Odinak	Sep 2010	B2
7809570	Kennewick	Oct 2010	B2
7818176	Freeman	Oct 2010	B2
7831426	Bennett	Nov 2010	B2
7831433	Belvin	Nov 2010	B1
7856358	Ho	Dec 2010	B2
7873519	Bennett	Jan 2011	B2
7873523	Potter	Jan 2011	B2
7873654	Bernard	Jan 2011	B2
7881936	Longe	Feb 2011	B2
7890324	Bangalore	Feb 2011	B2
7894849	Kass	Feb 2011	B2
7902969	Obradovich	Mar 2011	B2
7917367	DiCristo	Mar 2011	B2
7920682	Byrne	Apr 2011	B2
7949529	Weider	May 2011	B2
7949537	Walker	May 2011	B2
7953732	Frank	May 2011	B2
7974875	Quilici	Jul 2011	B1
7983917	Kennewick	Jul 2011	B2
7984287	Gopalakrishnan	Jul 2011	B2
8005683	Tessel	Aug 2011	B2
8015006	Kennewick	Sep 2011	B2
8024186	De Bonet	Sep 2011	B1
8027965	Takehara	Sep 2011	B2
8032383	Bhardwaj	Oct 2011	B1
8060367	Keaveney	Nov 2011	B2
8069046	Kennewick	Nov 2011	B2
8073681	Baldwin	Dec 2011	B2
8077975	Ma	Dec 2011	B2
8082153	Coffman	Dec 2011	B2
8086463	Ativanichayaphong	Dec 2011	B2
8103510	Sato	Jan 2012	B2
8112275	Kennewick	Feb 2012	B2
8140327	Kennewick	Mar 2012	B2
8140335	Kennewick	Mar 2012	B2
8145489	Freeman	Mar 2012	B2
8150694	Kennewick	Apr 2012	B2
8155962	Kennewick	Apr 2012	B2
8170867	Germain	May 2012	B2
8180037	Delker	May 2012	B1
8195468	Weider	Jun 2012	B2
8200485	Lee	Jun 2012	B1
8204751	Di Fabbrizio	Jun 2012	B1
8219399	Lutz	Jul 2012	B2
8219599	Tunstall-Pedoe	Jul 2012	B2
8224652	Wang	Jul 2012	B2
8255224	Singleton	Aug 2012	B2
8326599	Tomeh	Dec 2012	B2
8326627	Kennewick	Dec 2012	B2
8326634	DiCristo	Dec 2012	B2
8326637	Baldwin	Dec 2012	B2
8332224	DiCristo	Dec 2012	B2
8340975	Rosenberger	Dec 2012	B1
8346563	Hjelm	Jan 2013	B1
8370147	Kennewick	Feb 2013	B2
8447607	Weider	May 2013	B2
8447651	Scholl	May 2013	B1
8452598	Kennewick	May 2013	B2
8503995	Ramer	Aug 2013	B2
8509403	Chiu	Aug 2013	B2
8515765	Baldwin	Aug 2013	B2
8527274	Freeman	Sep 2013	B2
8577671	Barve	Nov 2013	B1
8589161	Kennewick	Nov 2013	B2
8612205	Hanneman	Dec 2013	B2
8612206	Chalabi	Dec 2013	B2
8620659	DiCristo	Dec 2013	B2
8719005	Lee	May 2014	B1
8719009	Baldwin	May 2014	B2
8719026	Kennewick	May 2014	B2
8731929	Kennewick	May 2014	B2
8738380	Baldwin	May 2014	B2
8849652	Weider	Sep 2014	B2
8849670	DiCristo	Sep 2014	B2
8849696	Pansari	Sep 2014	B2
8849791	Hertschuh	Sep 2014	B1
8886536	Freeman	Nov 2014	B2
8972243	Strom	Mar 2015	B1
8983839	Kennewick	Mar 2015	B2
9009046	Stewart	Apr 2015	B1
9015049	Baldwin	Apr 2015	B2
9037455	Faaborg	May 2015	B1
9070366	Mathias	Jun 2015	B1
9070367	Hoffmeister	Jun 2015	B1
9105266	Baldwin	Aug 2015	B2
9171541	Kennewick	Oct 2015	B2
9269097	Freeman	Feb 2016	B2
9305548	Kennewick	Apr 2016	B2
9308445	Merzenich	Apr 2016	B1
9318108	Gruber	Apr 2016	B2
9406078	Freeman	Aug 2016	B2
9443514	Taubman	Sep 2016	B1
9502025	Kennewick	Nov 2016	B2
20010039492	Nemoto	Nov 2001	A1
20010041980	Howard	Nov 2001	A1
20010047261	Kassan	Nov 2001	A1
20010049601	Kroeker	Dec 2001	A1
20010054087	Flom	Dec 2001	A1
20020002548	Roundtree	Jan 2002	A1
20020007267	Batchilo	Jan 2002	A1
20020010584	Schultz	Jan 2002	A1
20020015500	Belt	Feb 2002	A1
20020022927	Lemelson	Feb 2002	A1
20020022956	Ukrainczyk	Feb 2002	A1
20020029186	Roth	Mar 2002	A1
20020029261	Shibata	Mar 2002	A1
20020032752	Gold	Mar 2002	A1
20020035501	Handel	Mar 2002	A1
20020040297	Tsiao	Apr 2002	A1
20020049535	Rigo	Apr 2002	A1
20020049805	Yamada	Apr 2002	A1
20020059068	Rose	May 2002	A1
20020065568	Silfvast	May 2002	A1
20020067839	Heinrich	Jun 2002	A1
20020069059	Smith	Jun 2002	A1
20020069071	Knockeart	Jun 2002	A1
20020073176	Ikeda	Jun 2002	A1
20020082911	Dunn	Jun 2002	A1
20020087312	Lee	Jul 2002	A1
20020087326	Lee	Jul 2002	A1
20020087525	Abbott	Jul 2002	A1
20020107694	Lerg	Aug 2002	A1
20020120609	Lang	Aug 2002	A1
20020124050	Middeljans	Sep 2002	A1
20020133347	Schoneburg	Sep 2002	A1
20020133354	Ross	Sep 2002	A1
20020133402	Faber	Sep 2002	A1
20020135618	Maes	Sep 2002	A1
20020138248	Corston-Oliver	Sep 2002	A1
20020143532	McLean	Oct 2002	A1
20020143535	Kist	Oct 2002	A1
20020152260	Chen	Oct 2002	A1
20020161646	Gailey	Oct 2002	A1
20020161647	Gailey	Oct 2002	A1
20020169597	Fain	Nov 2002	A1
20020173333	Buchholz	Nov 2002	A1
20020173961	Guerra	Nov 2002	A1
20020184373	Maes	Dec 2002	A1
20020188602	Stubler	Dec 2002	A1
20020198714	Zhou	Dec 2002	A1
20030005033	Mohan	Jan 2003	A1
20030014261	Kageyama	Jan 2003	A1
20030016835	Elko	Jan 2003	A1
20030036903	Konopka	Feb 2003	A1
20030046071	Wyman	Mar 2003	A1
20030046281	Son	Mar 2003	A1
20030046346	Mumick	Mar 2003	A1
20030064709	Gailey	Apr 2003	A1
20030065427	Funk	Apr 2003	A1
20030069734	Everhart	Apr 2003	A1
20030069880	Harrison	Apr 2003	A1
20030088421	Maes	May 2003	A1
20030093419	Bangalore	May 2003	A1
20030097249	Walker	May 2003	A1
20030110037	Walker	Jun 2003	A1
20030112267	Belrose	Jun 2003	A1
20030115062	Walker	Jun 2003	A1
20030120493	Gupta	Jun 2003	A1
20030135488	Amir	Jul 2003	A1
20030144846	Denenberg	Jul 2003	A1
20030158731	Falcon	Aug 2003	A1
20030161448	Parolkar	Aug 2003	A1
20030167167	Gong	Sep 2003	A1
20030174155	Weng	Sep 2003	A1
20030182132	Niemoeller	Sep 2003	A1
20030187643	VanThong	Oct 2003	A1
20030204492	Wolf	Oct 2003	A1
20030206640	Malvar	Nov 2003	A1
20030212550	Ubale	Nov 2003	A1
20030212558	Matula	Nov 2003	A1
20030212562	Patel	Nov 2003	A1
20030225825	Healey	Dec 2003	A1
20030233230	Ammicht	Dec 2003	A1
20030236664	Sharma	Dec 2003	A1
20040006475	Ehlen	Jan 2004	A1
20040010358	Oesterling	Jan 2004	A1
20040025115	Sienel	Feb 2004	A1
20040030741	Wolton	Feb 2004	A1
20040036601	Obradovich	Feb 2004	A1
20040044516	Kennewick	Mar 2004	A1
20040093567	Schabes	May 2004	A1
20040098245	Walker	May 2004	A1
20040117179	Balasuriya	Jun 2004	A1
20040117804	Scahill	Jun 2004	A1
20040122673	Park	Jun 2004	A1
20040122674	Bangalore	Jun 2004	A1
20040133793	Ginter	Jul 2004	A1
20040140989	Papageorge	Jul 2004	A1
20040143440	Prasad	Jul 2004	A1
20040148154	Acero	Jul 2004	A1
20040148170	Acero	Jul 2004	A1
20040158555	Seedman	Aug 2004	A1
20040166832	Portman	Aug 2004	A1
20040167771	Duan	Aug 2004	A1
20040172247	Yoon	Sep 2004	A1
20040172258	Dominach	Sep 2004	A1
20040189697	Fukuoka	Sep 2004	A1
20040193408	Hunt	Sep 2004	A1
20040193420	Kennewick	Sep 2004	A1
20040199375	Ehsani	Oct 2004	A1
20040199389	Geiger	Oct 2004	A1
20040201607	Mulvey	Oct 2004	A1
20040205671	Sukehiro	Oct 2004	A1
20040243393	Wang	Dec 2004	A1
20040243417	Pitts	Dec 2004	A9
20040247092	Timmins	Dec 2004	A1
20040249636	Applebaum	Dec 2004	A1
20050015256	Kargman	Jan 2005	A1
20050021331	Huang	Jan 2005	A1
20050021334	Iwahashi	Jan 2005	A1
20050021470	Martin	Jan 2005	A1
20050021826	Kumar	Jan 2005	A1
20050033574	Kim	Feb 2005	A1
20050033582	Gadd	Feb 2005	A1
20050043940	Elder	Feb 2005	A1
20050080632	Endo	Apr 2005	A1
20050102282	Linden	May 2005	A1
20050114116	Fiedler	May 2005	A1
20050125232	Gadd	Jun 2005	A1
20050131673	Koizumi	Jun 2005	A1
20050137850	Odell	Jun 2005	A1
20050137877	Oesterling	Jun 2005	A1
20050143994	Mori	Jun 2005	A1
20050144013	Fujimoto	Jun 2005	A1
20050144187	Che	Jun 2005	A1
20050149319	Honda	Jul 2005	A1
20050216254	Gupta	Sep 2005	A1
20050222763	Uyeki	Oct 2005	A1
20050234637	Obradovich	Oct 2005	A1
20050234727	Chiu	Oct 2005	A1
20050246174	DeGolia	Nov 2005	A1
20050283364	Longe	Dec 2005	A1
20050283532	Kim	Dec 2005	A1
20050283752	Fruchter	Dec 2005	A1
20060041431	Maes	Feb 2006	A1
20060046740	Johnson	Mar 2006	A1
20060047509	Ding	Mar 2006	A1
20060072738	Louis	Apr 2006	A1
20060074670	Weng	Apr 2006	A1
20060074671	Farmaner	Apr 2006	A1
20060080098	Campbell	Apr 2006	A1
20060100851	Schonebeck	May 2006	A1
20060106769	Gibbs	May 2006	A1
20060129409	Mizutani	Jun 2006	A1
20060130002	Hirayama	Jun 2006	A1
20060182085	Sweeney	Aug 2006	A1
20060206310	Ravikumar	Sep 2006	A1
20060217133	Christenson	Sep 2006	A1
20060236343	Chang	Oct 2006	A1
20060242017	Libes	Oct 2006	A1
20060253247	de Silva	Nov 2006	A1
20060253281	Letzt	Nov 2006	A1
20060285662	Yin	Dec 2006	A1
20070011159	Hillis	Jan 2007	A1
20070033005	Di Cristo	Feb 2007	A1
20070033020	Francois	Feb 2007	A1
20070033526	Thompson	Feb 2007	A1
20070038436	Cristo	Feb 2007	A1
20070038445	Helbing	Feb 2007	A1
20070043569	Potter	Feb 2007	A1
20070043574	Coffman	Feb 2007	A1
20070043868	Kumar	Feb 2007	A1
20070050191	Weider	Mar 2007	A1
20070050279	Huang	Mar 2007	A1
20070055525	Kennewick	Mar 2007	A1
20070061067	Zeinstra	Mar 2007	A1
20070061735	Hoffberg	Mar 2007	A1
20070073544	Millett	Mar 2007	A1
20070078708	Yu	Apr 2007	A1
20070078709	Rajaram	Apr 2007	A1
20070078814	Flowers	Apr 2007	A1
20070094003	Huang	Apr 2007	A1
20070100797	Thun	May 2007	A1
20070112555	Lavi	May 2007	A1
20070112630	Lau	May 2007	A1
20070118357	Kasravi	May 2007	A1
20070124057	Prieto	May 2007	A1
20070135101	Ramati	Jun 2007	A1
20070146833	Satomi	Jun 2007	A1
20070162296	Altberg	Jul 2007	A1
20070174258	Jones	Jul 2007	A1
20070179778	Gong	Aug 2007	A1
20070185859	Flowers	Aug 2007	A1
20070186165	Maislos	Aug 2007	A1
20070192309	Fischer	Aug 2007	A1
20070198267	Jones	Aug 2007	A1
20070203699	Nagashima	Aug 2007	A1
20070203736	Ashton	Aug 2007	A1
20070208732	Flowers	Sep 2007	A1
20070214182	Rosenberg	Sep 2007	A1
20070250901	McIntire	Oct 2007	A1
20070265850	Kennewick	Nov 2007	A1
20070266257	Camaisa	Nov 2007	A1
20070276651	Bliss	Nov 2007	A1
20070294615	Sathe	Dec 2007	A1
20070299824	Pan	Dec 2007	A1
20080014908	Vasant	Jan 2008	A1
20080034032	Healey	Feb 2008	A1
20080046311	Shahine	Feb 2008	A1
20080059188	Konopka	Mar 2008	A1
20080065386	Cross	Mar 2008	A1
20080065389	Cross	Mar 2008	A1
20080065390	Ativanichayaphong	Mar 2008	A1
20080086455	Meisels	Apr 2008	A1
20080091406	Baldwin	Apr 2008	A1
20080103761	Printz	May 2008	A1
20080103781	Wasson	May 2008	A1
20080104071	Pragada	May 2008	A1
20080109285	Reuther	May 2008	A1
20080115163	Gilboa	May 2008	A1
20080126091	Clark	May 2008	A1
20080133215	Sarukkai	Jun 2008	A1
20080140385	Mahajan	Jun 2008	A1
20080147396	Wang	Jun 2008	A1
20080147410	Odinak	Jun 2008	A1
20080147637	Li	Jun 2008	A1
20080154604	Sathish	Jun 2008	A1
20080162471	Bernard	Jul 2008	A1
20080177530	Cross	Jul 2008	A1
20080184164	Di Fabbrizio	Jul 2008	A1
20080189110	Freeman	Aug 2008	A1
20080228496	Yu	Sep 2008	A1
20080235023	Kennewick	Sep 2008	A1
20080235027	Cross	Sep 2008	A1
20080269958	Filev	Oct 2008	A1
20080270135	Goel	Oct 2008	A1
20080270224	Portman	Oct 2008	A1
20080294437	Nakano	Nov 2008	A1
20080294994	Kruger	Nov 2008	A1
20080306743	Di Fabbrizio	Dec 2008	A1
20080319751	Kennewick	Dec 2008	A1
20090006077	Keaveney	Jan 2009	A1
20090006194	Sridharan	Jan 2009	A1
20090018829	Kuperstein	Jan 2009	A1
20090024476	Baar	Jan 2009	A1
20090030686	Weng	Jan 2009	A1
20090052635	Jones	Feb 2009	A1
20090055176	Hu	Feb 2009	A1
20090067599	Agarwal	Mar 2009	A1
20090076827	Bulitta	Mar 2009	A1
20090106029	DeLine	Apr 2009	A1
20090117885	Roth	May 2009	A1
20090144131	Chiu	Jun 2009	A1
20090144271	Richardson	Jun 2009	A1
20090150156	Kennewick	Jun 2009	A1
20090157382	Bar	Jun 2009	A1
20090164216	Chengalvarayan	Jun 2009	A1
20090171664	Kennewick	Jul 2009	A1
20090171912	Nash	Jul 2009	A1
20090197582	Lewis	Aug 2009	A1
20090216540	Tessel	Aug 2009	A1
20090248565	Chuang	Oct 2009	A1
20090248605	Mitchell	Oct 2009	A1
20090259561	Boys	Oct 2009	A1
20090259646	Fujita	Oct 2009	A1
20090265163	Li	Oct 2009	A1
20090271194	Davis	Oct 2009	A1
20090273563	Pryor	Nov 2009	A1
20090276700	Anderson	Nov 2009	A1
20090287680	Paek	Nov 2009	A1
20090299745	Kennewick	Dec 2009	A1
20090299857	Brubaker	Dec 2009	A1
20090304161	Pettyjohn	Dec 2009	A1
20090307031	Winkler	Dec 2009	A1
20090313026	Coffman	Dec 2009	A1
20090319517	Guha	Dec 2009	A1
20100023320	Cristo	Jan 2010	A1
20100023331	Duta	Jan 2010	A1
20100029261	Mikkelsen	Feb 2010	A1
20100036967	Caine	Feb 2010	A1
20100049501	Kennewick	Feb 2010	A1
20100049514	Kennewick	Feb 2010	A1
20100057443	Cristo	Mar 2010	A1
20100063880	Atsmon	Mar 2010	A1
20100064025	Nelimarkka	Mar 2010	A1
20100094707	Freer	Apr 2010	A1
20100138300	Wallis	Jun 2010	A1
20100145700	Kennewick	Jun 2010	A1
20100185512	Borger	Jul 2010	A1
20100204986	Kennewick	Aug 2010	A1
20100204994	Kennewick	Aug 2010	A1
20100217604	Baldwin	Aug 2010	A1
20100268536	Suendermann	Oct 2010	A1
20100286985	Kennewick	Nov 2010	A1
20100299142	Freeman	Nov 2010	A1
20100312566	Odinak	Dec 2010	A1
20100318357	Istvan	Dec 2010	A1
20100331064	Michelstein	Dec 2010	A1
20110022393	Waller	Jan 2011	A1
20110106527	Chiu	May 2011	A1
20110112827	Kennewick	May 2011	A1
20110112921	Kennewick	May 2011	A1
20110119049	Ylonen	May 2011	A1
20110131036	DiCristo	Jun 2011	A1
20110131045	Cristo	Jun 2011	A1
20110231182	Weider	Sep 2011	A1
20110231188	Kennewick	Sep 2011	A1
20110238409	Larcheveque	Sep 2011	A1
20110307167	Taschereau	Dec 2011	A1
20120022857	Baldwin	Jan 2012	A1
20120041753	Dymetman	Feb 2012	A1
20120046935	Nagao	Feb 2012	A1
20120101809	Kennewick	Apr 2012	A1
20120101810	Kennewick	Apr 2012	A1
20120109753	Kennewick	May 2012	A1
20120150620	Mandyam	Jun 2012	A1
20120150636	Freeman	Jun 2012	A1
20120239498	Ramer	Sep 2012	A1
20120240060	Pennington	Sep 2012	A1
20120265528	Gruber	Oct 2012	A1
20120278073	Weider	Nov 2012	A1
20130006734	Ocko	Jan 2013	A1
20130054228	Baldwin	Feb 2013	A1
20130060625	Davis	Mar 2013	A1
20130080177	Chen	Mar 2013	A1
20130211710	Kennewick	Aug 2013	A1
20130253929	Weider	Sep 2013	A1
20130254314	Chow	Sep 2013	A1
20130297293	Cristo	Nov 2013	A1
20130304473	Baldwin	Nov 2013	A1
20130311324	Stoll	Nov 2013	A1
20130332454	Stuhec	Dec 2013	A1
20130339022	Baldwin	Dec 2013	A1
20140006951	Hunter	Jan 2014	A1
20140012577	Freeman	Jan 2014	A1
20140025371	Min	Jan 2014	A1
20140108013	Cristo	Apr 2014	A1
20140156278	Kennewick	Jun 2014	A1
20140195238	Terao	Jul 2014	A1
20140236575	Tur	Aug 2014	A1
20140249821	Kennewick	Sep 2014	A1
20140249822	Baldwin	Sep 2014	A1
20140278413	Pitschel	Sep 2014	A1
20140278416	Schuster	Sep 2014	A1
20140288934	Kennewick	Sep 2014	A1
20140330552	Bangalore	Nov 2014	A1
20140365222	Weider	Dec 2014	A1
20150019211	Simard	Jan 2015	A1
20150019217	Cristo	Jan 2015	A1
20150019227	Anandarajah	Jan 2015	A1
20150066479	Pasupalak	Mar 2015	A1
20150066627	Freeman	Mar 2015	A1
20150073910	Kennewick	Mar 2015	A1
20150095159	Kennewick	Apr 2015	A1
20150142447	Kennewick	May 2015	A1
20150170641	Kennewick	Jun 2015	A1
20150193379	Mehta	Jul 2015	A1
20150199339	Mirkin	Jul 2015	A1
20150228276	Baldwin	Aug 2015	A1
20150293917	Bufe	Oct 2015	A1
20150348544	Baldwin	Dec 2015	A1
20150348551	Gruber	Dec 2015	A1
20150364133	Freeman	Dec 2015	A1
20160049152	Kennewick	Feb 2016	A1
20160078482	Kennewick	Mar 2016	A1
20160078491	Kennewick	Mar 2016	A1
20160078504	Kennewick	Mar 2016	A1
20160078773	Carter	Mar 2016	A1
20160110347	Kennewick	Apr 2016	A1
20160148610	Kennewick	May 2016	A1
20160148612	Guo	May 2016	A1
20160188292	Carter	Jun 2016	A1
20160188573	Tang	Jun 2016	A1
20160217785	Kennewick	Jul 2016	A1
20160335676	Freeman	Nov 2016	A1

Foreign Referenced Citations (35)

Number	Date	Country
1433554	Jul 2003	CN
1860496	Nov 2006	CN
1320043	Jun 2003	EP
1646037	Apr 2006	EP
H08263258	Oct 1996	JP
H11249773	Sep 1999	JP
2001071289	Mar 2001	JP
2006146881	Jun 2006	JP
2008027454	Feb 2008	JP
2008058465	Mar 2008	JP
2008139928	Jun 2008	JP
2011504304	Feb 2011	JP
2012518847	Aug 2012	JP
9946763	Sep 1999	WO
0021232	Jan 2000	WO
0046792	Jan 2000	WO
0171609	Sep 2001	WO
0178065	Oct 2001	WO
2004072954	Aug 2004	WO
2005010702	Feb 2005	WO
2007019318	Jan 2007	WO
2007021587	Jan 2007	WO
2007027546	Jan 2007	WO
2007027989	Jan 2007	WO
2008098039	Jan 2008	WO
2008118195	Jan 2008	WO
2009075912	Jan 2009	WO
2009145796	Jan 2009	WO
2009111721	Sep 2009	WO
2010096752	Jan 2010	WO
2016044290	Mar 2016	WO
2016044316	Mar 2016	WO
2016044319	Mar 2016	WO
2016044321	Mar 2016	WO
2016061309	Apr 2016	WO

Non-Patent Literature Citations (23)

Entry
“Statement in Accordance with the Notice from the European Patent Office” dated Oct. 1, 2007 Concerning Business Methods (OJ EPO Nov. 2007, 592-593), XP002456252.
Arrington, Michael, “Google Redefines GPS Navigation Landscape: Google Maps Navigation for Android 2.0”, TechCrunch, printed from the Internet <http://www.techcrunch.com/2009/10/28/google-redefines-car-gps-navigation-google-maps-navigation-android/>, Oct. 28, 2009, 4 pages.
Bazzi, Issam et al., “Heterogeneous Lexical Units for Automatic Speech Recognition: Preliminary Investigations”, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 3, Jun. 5-9, 2000, XP010507574, pp. 1257-1260.
Belvin, Robert, et al., “Development of the HRL Route Navigation Dialogue System”, Proceedings of the First International Conference on Human Language Technology Research, San Diego, 2001, pp. 1-5.
Chai et al., “MIND: A Semantics-Based Multimodal Interpretation Framework for Conversational Systems”, Proceedings of the International CLASS Workshop on Natural, Intelligent and Effective Interaction in Multimodal Dialogue Systems, Jun. 2002, pp. 37-46.
Cheyer et al., “Multimodal Maps: An Agent-Based Approach”, International Conference on Cooperative Multimodal Communication (CMC/95), May 24-26, 1995, pp. 111-121.
Davis, Z., et al., A Personal Handheld Multi-Modal Shopping Assistant, IEEE, 2006, 9 pages.
El Meliani et al., “A Syllabic-Filler-Based Continuous Speech Recognizer for Unlimited Vocabulary”, Canadian Conference on Electrical and Computer Engineering, vol. 2, Sep. 5-8, 1995, pp. 1007-1010.
Elio et al., “On Abstract Task Models and Conversation Policies” in Workshop on Specifying and Implementing Conversation Policies, Autonomous Agents '99, Seattle, 1999, 10 pages.
Kirchhoff, Katrin, “Syllable-Level Desynchronisation of Phonetic Features for Speech Recognition”, Proceedings of the Fourth International Conference on Spoken Language, 1996, ICSLP 96, vol. 4, IEEE, 1996, 3 pages.
Kuhn, Thomas, et al., “Hybrid In-Car Speech Recognition for Mobile Multimedia Applications”, Vehicular Technology Conference, IEEE, Jul. 1999, pp. 2009-2013.
Lin, Bor-shen, et al., “A Distributed Architecture for Cooperative Spoken Dialogue Agents with Coherent Dialogue State and History”, ASRU'99, 1999, 4 pages.
Lind, R., et al., “The Network Vehicle—A Glimpse into the Future of Mobile Multi-Media”, IEEE Aerosp. Electron. Systems Magazine, vol. 14, No. 9, Sep. 1999, pp. 27-32.
Mao, Mark Z., “Automatic Training Set Segmentation for Multi-Pass Speech Recognition”, Department of Electrical Engineering, Stanford University, CA, copyright 2005, IEEE, pp. I-685 to I-688.
O'Shaughnessy, Douglas, “Interacting with Computers by Voice: Automatic Speech Recognition and Synthesis”, Proceedings of the IEEE, vol. 91, No. 9, Sep. 1, 2003, XP011100665. pp. 1272-1305.
Reuters, “IBM to Enable Honda Drivers to Talk to Cars”, Charles Schwab & Co., Inc., Jul. 28, 2002, 1 page.
Turunen, “Adaptive Interaction Methods in Speech User Interfaces”, Conference on Human Factors in Computing Systems, Seattle, Washington, 2001, pp. 91-92.
Vanhoucke, Vincent, “Confidence Scoring and Rejection Using Multi-Pass Speech Recognition”, Nuance Communications, Menlo Park, CA, 2005, 4 pages.
Weng, Fuliang, et al., “Efficient Lattice Representation and Generation”, Speech Technology and Research Laboratory, SRI International, Menlo Park, CA, 1998, 4 pages.
Wu, Su-Lin, et al., “Incorporating Information from Syllable-Length Time Scales into Automatic Speech Recognition”, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, 1998, vol. 2, IEEE, 1998, 4 pages.
Wu, Su-Lin, et al., “Integrating Syllable Boundary Information into Speech Recognition”, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP-97, 1997, vol. 2, IEEE, 1997, 4 pages.
Zhao, Yilin, “Telematics: Safe and Fun Driving”, IEEE Intelligent Systems, vol. 17, Issue 1, 2002, pp. 10-14.
International Preliminary Report on Patentability dated Feb. 7, 2019 from corresponding International Patent Application No. PCT/US2017/044610, 7 pages.

Related Publications (1)

	Number	Date	Country
	20180032503 A1	Feb 2018	US

Provisional Applications (1)

	Number	Date	Country
	62368975	Jul 2016	US

System and method of disambiguating natural language processing requests

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

Field of Search

CPC

International Classifications

Abstract