AGGREGATING PERSONALIZED SUGGESTIONS FROM MULTIPLE SOURCES

Abstract

A computer provides a search interface that accesses multiple search providers, and aggregates their suggestions, providing a single, unified suggestion view across the multiple search providers. Suggestions are received from the multiple sources, such as a search engine on the internet or other public resource, and a search tool on the computer that accesses local or private resources. The suggestions are combined, ranked and displayed as a list to the user, from which the user is able to select.

Description

BACKGROUND

Many user interfaces for text-based search providers on a computer or on the internet provide one or more suggestions to a user while the user is entering a query for a search. Such suggestions typically are words or phrases that match and/or complete a string of characters that the user has entered. As the user enters more characters, the suggestions change to continue matching or completing the string already entered. A user can stop entering characters and select one of the suggested words or phrases as the query.

A search provider typically derives suggestions from history data and potential search results for the query as currently entered. For example, if a user's computer indexes files, then a search tool on that computer may provide suggestions based on files that match the query as entered. The search tool also may provide suggestions based on prior queries performed by the user using the search tool. Similarly, a search engine on the internet may provide suggestions based on resources, such as web pages, that match the query as entered. The search engine also may provide suggestions based on prior queries of that user and/or other users using the search tool. In general, each search provider provides suggestions based on the history or indexed data associated with that search provider.

SUMMARY

This Summary introduces selected concepts in simplified form that are further described below in the Detailed Description. This Summary is intended neither to identify key or essential features of the claimed subject matter, nor to limit the scope of the claimed subject matter.

A computer provides a search interface that accesses multiple search providers, and aggregates their suggestions, providing a single, unified suggestion view across the multiple search providers. Suggestions are received from the multiple sources, such as a search engine on the internet or other public resource, and a search tool on the computer that accesses local or private resources. The suggestions are combined, ranked and displayed as a list to the user, from which the user is able to select.

In the following description, reference is made to the accompanying drawings which form a part hereof, and in which are shown, by way of illustration, specific example implementations of this technique. It is understood that other embodiments may be utilized and structural changes may be made without departing from the scope of the disclosure.

DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of an example system showing multiple users accessing multiple search providers providing suggestions.

FIG. 2 is a data flow diagram illustrating an example implementation of a search interface that unifies suggestions from multiple search providers.

FIG. 3 is a flow chart an example implementation of combining suggestions.

FIG. 4 is an illustration of an example graphical user interface illustrating combined suggestions.

FIG. 5 is a data flow diagram describing an example implementation of a suggestion system.

FIG. 6 is a flow chart describing an example implementation of a suggestion system.

FIG. 7 is a block diagram of an example computing device in which such a system can be implemented.

DETAILED DESCRIPTION

The following section provides an example operating environment in which aggregated suggestions from multiple sources can be implemented.

Referring to FIG. 1, a system 100 includes a first computer 102 and a second computer 104. The computers 102, 104 are connected through a computer network 106 to one or more other computers, such as a website that provides a search engine 108. The first and second computers also may connect to computers that provide a storage service 110, which provides remote storage of data files. First and second computers also typically have local storage 112, 114 on which data files are stored through a file system of an operating system of the computers 102, 104. A typical computer configuration is described in more detail below in connection with FIG. 7.

In such a configuration, a user of a computer 102 (or 104) can search for information in data files in local storage 112, the storage service 110 or on the search engine 108. Accessing each of these resources generally involves a different search provider—a search tool on the local computer for the local storage, a search interface for the storage service, and a search interface for the search engine. As shown in FIG. 1, if these search providers provide suggestions, an initial query is sent to the search provider used by the user, which responds with one or more suggestions.

A search interface (not shown) in the computers receives user input to provide an initial query and provide the initial query to the search providers. The search interface receives the suggestions from the search providers, and combines them into a single, unified list 136, 138 of suggestions which is displayed on a display 140, 142 for the user. The user can select from among these suggestions.

Given this context, an example implementation will be described in more detail in connection with FIGS. 2-6.

In FIG. 2, a data flow diagram of a search interface 200 will now be provided. Such a search interface can be implemented on the user's computer or on a remote service accessed through the user's computer over a computer network. In either case, the search interface receives suggestions from other search providers over the computer network, and can receive suggestions from search providers on the same computer as the search interface.

The search interface 200 includes a user interaction module 202 that receives input 204 from and provides output 206 to the user. The user input generally includes alphanumeric characters entered by the user through an input device (not shown) and user selections in response to query suggestions. The output includes data to be displayed or otherwise communicated to the user, such as an interactive display of the user input as the user enters a query and suggestions provided to the user. In turn, the user interaction module 202 provides query terms 208 and user selections 210 to other modules in the system, and receives query results 212 and suggestions 214 from other modules in the system.

A query processing module 220 receives the current query terms 208 and distributes them as queries 222 to multiple search providers 230. Each search provider 230 provides suggestions 224 in reply. Query processing module combines the suggestions 224, in a manner described below, into the suggestions 214 provided to the user. When the user selection 210 indicates that the query is complete, the query processing module sends the search terms 208 to one or more of the multiple search providers 230 as queries, and receives query results 226 in reply, from which query results 212 are provided to the user.

The query processing module also can track queries, suggestions and query results in history data 240. History data can be stored in local storage (not shown) or remote storage (also not shown). In remote storage, such history data can be aggregated for the same user across multiple devices, and with similar history data from other users as well.

Referring now to FIG. 3, an example implementation of a process for combining suggestion results from multiple sources will now be described.

In this implementation, the combination of suggestions involves ordering suggestions first by ranking the search providers from which the suggestion originate, and then by ranking suggestions within that grouping. Duplicates can be removed. Clearly, other implementations are possible, such as scoring and ranking each suggestion independently of its search provider.

In this example implementation, each search provider is assigned a set of weights to be applied to a set of features for a suggestion for that search provider. There can be an arbitrary set of features for each provider. Thus each provider can have different features, a different number of features and different weightings for those features. In this example, the features for a suggestion are related to the nature of the match between the query terms and the item resulting in the suggestion. For example, a match can be an exact match, a partial match or regular expression match, each of which can be represented by a binary value, e.g., 0 or 1. Thus, any particular suggestion has a related set of features {[0,1], [0,1], [0,1]}, respectively for exact match, regular expression match and partial match, indicating the nature of the match between that suggestion and the query terms. For each search provider, a weight is assigned to each feature in its corresponding set of features.

As an example, for files or other local search, the set of weights can be: {0.8, 0.6, 0.2}, respectively, for exact match, regular expression match and partial match. Thus, the file search facility can return, as a suggestion, a file name that is a partial match, thus having features {0,0,1}, which results in a score for that suggestion of 0.8*0+0.6*0+0.2*1=0.2.

As another example, for the web or other non-local search, a different set of weights can be used, such as: {0.6, 0.4, 0.2}, respectively, for exact match, regular expression match and partial match. Thus, the web search provider can return, as a suggestion, a document that is an exact match, thus having features {1,0,0}, which results in a score for that suggestion of 0.6*1+0.4*0+0.2*0=0.6.

The scores for the suggestions from the search providers can be used as an “order score” by which the search providers are ordered.

If multiple suggestions are provided by a search provider, then an average of a number of these suggestions can be computed and this average can be used as an “order score” for that search provider. Alternatively an order score or weight can be computed for or assigned to the search provider. As noted below, using history data gathered over the computer network related to suggestions, weights can be determined from the web-based data even for local search providers, such as files and settings.

If the order score for a search provider is less than a designated threshold, then the that search provider can be excluded from, hidden or collapsed in the set of suggested presented to the user. A value can be stored for each search provider and the display behavior can be dependent on this value.

This example implementation is reflected in the flow chart shown in FIG. 3. The suggestion engine submits 300 the current query to local search providers, and submits 302 the query to remote search providers. Suggestions from the local applications and remote applications are received 304. For each search provider, an order score is computed 306. Given the order score and the suggestions for each search provider, the suggestion engine orders and displays 308 the received suggestions.

An example graphical user interface 400 for displaying such suggestions is shown in FIG. 4. The interface includes a data entry slot 402 for the user to enter a current query. The current query is interactively displayed as characters, words or phrases are entered by the user. Suggestions are shown below the query at 404. An icon 406 can be used to indicate the type of suggestion. Other data, such as a file name or resource name 408, can be displayed adjacent the icon 406. Different colors or other identifying indicia can be associated with different types of search providers, so that the user can readily distinguish suggestions from different sources.

In the example shown in FIG. 4, a large number of different sources of suggestions are used. In this example, the first items are from a history of local commands. Those are followed by an application suggestion, a music selection, a settings suggestion, file suggestions, web site suggestions, contact suggestions, search history suggestions, and finally search suggestions.

Given the displayed list of suggestions, a user can either continue to enter additional characters, words or phrases, or select from among the suggestions displayed.

The information about the displayed suggestions, and any user selection, can be tracked and later used to provide ways to improve the weighting, scoring or other aspects of the suggestion engine. This information can also be aggregated over multiple users.

Referring now to FIG. 5, a dataflow diagram for an example implementation of a centralized system for aggregating such data from multiple users will now be described. Such a system includes what is called herein a “suggestion service” 500. This suggestion service can be used, for example, by a web-based search engine 502, which can be one of the search providers accessed through the search interface of FIGS. 1 and 2 to provide suggestions 503.

The suggestion service is a server computer running a computer program and connected to a computer network (not shown) over which user computers 504 send user suggestion data 506. The user suggestion data can include any suggestions presented to the user, the query that initiated the suggestions, and any selection made by the user from among the suggestions. Locally stored history data can include file names or other specific data about the specific resource accessed by a user after selecting a suggestion. When providing data to the suggestion service, the information provided to the service can include the features and weightings of the suggestions and search providers, and the features, weightings and identity of the search provider. Given the user suggestion data captured by the suggestion service, various analyses can be performed on the data to attempt to improve the suggestions provided by the search engine 502 and/or to improve the weightings or other techniques used to rank suggestions.

In particular, a machine learning algorithm can be used to process the features, weightings and search providers presented to and selected by the user. This information can be processed to optimize a set of weights such that the features and weightings presented to the user tend to place the selected search provider at the top of the list of suggestions presented to the user. In this way, the features and weights determined can be said to result in an ordering of suggestions that is close to what the user might expect. The features and weights determined for each search provider by the suggestion service can be passed to a computer which uses them to aggregate and present suggestions to the user of that computer.

Referring now to FIG. 6, a flow chart illustrating an example operation of the system of FIG. 5 will now be described.

For such a service that captures user suggestion data, the service first has the user log in, as noted at 600. Alternatively, data can be collected anonymously across several users. A user then initiates a query and receives suggestions, including suggestions from the search engine associated with the suggestion service. Information about the received suggestions, along with information about their relative ordering when presented to the user, is transmitted to the suggestion service, in response to which the suggestion service receives 602 the suggestion data and stores 604 the suggestion data. In response to a user selecting one of the suggestions, the user computer transmits data about the selected suggestion to the suggestion service, which in turn receives 606 the selection and stores 608 data about the suggestion. The stored data then can be analyzed 610 to improve the suggestion process. Further, the system with the stored data can act as a search provider that provides an aggregate search history across multiple search providers. This search provider can provide suggestions based on a user's history with the search engine, the user's aggregate history over multiple devices and multiple search providers, and/or based on the histories stored for other users determined to be similar to the user, which have a likelihood of being relevant to the user.

Having now described an example implementation, a computer with which components of such a system are designed to operate will now be described. The following description is intended to provide a brief, general description of a suitable computer with which such a system can be implemented. The computer can be any of a variety of general purpose or special purpose computing hardware configurations. Examples of well-known computers that may be suitable include, but are not limited to, personal computers, server computers, hand-held or laptop devices (for example, media players, notebook computers, cellular phones, personal data assistants, voice recorders), multiprocessor systems, microprocessor-based systems, set top boxes, game consoles, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

FIG. 7 illustrates an example of a suitable computer. This is only one example of a suitable computer and is not intended to suggest any limitation as to the scope of use or functionality of such a computer.

With reference to FIG. 7, an example computer 700, in a basic configuration, includes at least one processing unit 706 and memory 704. The computer may include multiple processing units and/or additional co-processing units such as graphics processing unit 720. Depending on the exact configuration and type of computer, memory 704 may be volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.) or some combination of the two. This configuration is illustrated in FIG. 7 by dashed line 706.

Additionally, computer 700 may also have additional features/functionality. For example, computer 700 may also include additional storage (removable and/or non-removable) including, but not limited to, magnetic or optical disks or tape. Such additional storage is illustrated in FIG. 7 by removable storage 708 and non-removable storage 710. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer program instructions, data structures, program modules or other data. Memory 704, removable storage 708 and non-removable storage 710 are all examples of computer storage media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by computer 700. Any such computer storage media may be part of computer 700.

Computer 700 may also contain communications connection(s) 712 that allow the device to communicate with other devices over a communication medium. Communication media typically carry computer program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and include any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal, thereby changing the configuration or state of the receiving device of the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Communications connections 712 are devices that interface with the communication media to transmit data over and receive data from communication media, such as a network interface.

Computer 700 may have various input device(s) 714 such as a keyboard, mouse, pen, camera, touch input device, and so on. Output device(s) 716 such as a display, speakers, a printer, and so on may also be included. All of these devices are well known in the art and need not be discussed at length here. Various input and output devices can implement a natural user interface (NUI), which is any interface technology that enables a user to interact with a device in a “natural” manner, free from artificial constraints imposed by input devices such as mice, keyboards, remote controls, and the like.

Examples of NUI methods include those relying on speech recognition, touch and stylus recognition, gesture recognition both on screen and adjacent to the screen, air gestures, head and eye tracking, voice and speech, vision, touch, gestures, and machine intelligence, and may include the use of touch sensitive displays, voice and speech recognition, intention and goal understanding, motion gesture detection using depth cameras (such as stereoscopic camera systems, infrared camera systems, and other camera systems and combinations of these), motion gesture detection using accelerometers or gyroscopes, facial recognition, three dimensional displays, head, eye, and gaze tracking, immersive augmented reality and virtual reality systems, all of which provide a more natural interface, as well as technologies for sensing brain activity using electric field sensing electrodes (EEG and related methods).

Each component of this system that operates on a computer generally is implemented by software, such as one or more computer programs, which include computer-executable instructions and/or computer-interpreted instructions, such as program modules, being processed by the computer. Generally, program modules include routines, programs, objects, components, data structures, and so on, that, when processed by a processing unit, instruct the processing unit to perform particular tasks or implement particular abstract data types. This computer system enforces licensing restrictions may be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

Alternatively, or in addition, the functionally described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include Field-programmable Gate Arrays (FPGAs), Program-specific Integrated Circuits (ASICs), Program-specific Standard Products (ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), etc.

The terms “article of manufacture”, “process”, “machine” and “composition of matter” in the preambles of the appended claims are intended to limit the claims to subject matter deemed to fall within the scope of patentable subject matter defined by the use of these terms in 35 U.S.C. § 101.

Any or all of the aforementioned alternate embodiments described herein may be used in any combination desired to form additional hybrid embodiments. It should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific implementations described above. The specific implementations described above are disclosed as examples only.

Claims

1. A method performed by one more computing devices, the one or more computing devices comprising respective processing hardware and storage hardware, the method comprising: displaying a query input area, receiving a query string inputted therein, and sending the query string to a plurality of search providers;receiving, from each search provider, a respective resource suggestion set, each resource suggestion set comprising a respective set of resource suggestions, each resource suggestion comprising a resource indication and a resource name for a corresponding resource;for each received suggestion set, ranking the resource suggestions therein relative to each other;ranking the resource suggestion sets relative to each other;displaying suggestion set representations, wherein each suggestion set representation comprises resource names of a corresponding suggestion set, wherein each suggestion set representation's resource names are displayed contiguously within the suggestion set representation according to the rankings of the corresponding resource suggestions, wherein the suggestion set representations are displayed in order relative to each other according to the rankings of the corresponding suggestion sets, and wherein each suggestion set representation comprises resource names of only the corresponding suggestion set; andresponding to a user input selecting a displayed resource name by using the corresponding resource indication to access the corresponding resource.

2. A method according to claim 1, wherein the method is performed by a single computing device.

3. A method according to claim 2, wherein one search provider executes on the single computing device, and wherein another search provider comprises a service accessed via a data network.

4. A method according to claim 1, wherein each resource suggestion further comprises respective features, and wherein the ranking the suggestion sets comprises computing a score of a suggestion set according to the features of the suggestions in the suggestion set.

5. A method according to claim 4, wherein the ranking the resource suggestions comprises computing respective scores according to the features thereof, and wherein the method further comprises omitting one or more resource names from the suggestion set representations according to the scores.

6. A method according to claim 1, further comprising computing scores for the suggestion sets and determining not to display a suggestion set representation for a suggestion set based on the score of the suggestion set.

7. A method according to claim 1, wherein a first computing device performs the ranking the resource suggestions and the ranking the suggestion sets, and wherein a second computing device performs the displaying the query input area and the displaying the suggestion set representations.

8. A method according to claim 1, wherein the displaying comprises graphically distinguishing the suggestion set representations from each other.

9. A computing device comprising: a display;processing hardware;storage hardware storing instructions that, when executed by the processing hardware, perform a process, the process comprising: displaying, by the display, a query input area, and receiving a query string inputted by a user in the query input area;sending the query string over a network to a query processing service, and receiving therefrom a first suggestion set, the first suggestion set comprising first suggestions, each first suggestion comprising a respective first name and first resource identifier;providing the query string to a local search provider executing on the computing device, and receiving therefrom a second suggestion set, the second suggestion set comprising second suggestions, each second suggestion comprising a respective second name and second resource identifier;obtaining rankings of the first suggestions relative to each other;obtaining rankings of the second suggestions relative to each other;obtaining rankings of the first and second suggestion sets relative to each other;displaying a first representation for the first suggestion set and a second representation for the second suggestion set, wherein the first and second representations are displayed in an order determined by the rankings of the first and suggestion sets, wherein the first representation comprises first names displayed in adjacent order according to the rankings of the first suggestions, wherein the second representation comprises second names displayed in adjacent order according to the rankings of the second suggestions, and wherein the displayed first names and second names are user-selectable; andresponding to a user input selecting one of the displayed first or second names by accessing and displaying the first or second resource corresponding thereto.

10. A computing device according to claim 9, the process further comprising ranking the first suggestions by computing scores for the respective features thereof, and omitting a first name from the first representation according to the score thereof.

11. A computing device according to claim 9, wherein the second search provider comprises, or interfaces with, a filesystem on the computing device, and wherein the second names comprise filenames of files in the filesystem.

12. A computing device according to claim 9, wherein the rankings of the first suggestions, the second suggestions, and the first and second suggestion sets are obtained via the network from a network service.

13. A computing device according to claim 9, wherein the rankings of the first suggestions are obtained by computing scores from features thereof, and wherein the rankings of the second suggestions are obtained by computing scores from features thereof.

14. A computing device according to claim 9, the process further comprising sending an indication of the selected first or second name to a network service that maintains a history of user selections of suggestion names from other computing devices.

15. A computing device according to claim 9, wherein the rankings are based at least in part on the history.

16. Computer-readable storage hardware, not comprising a signal, storing in formation that, when executed by a computing device, perform a process comprising: receiving a query string inputted by a user in a query input area;sending the query string over a network to a query service, and receiving therefrom a first suggestion set, the first suggestion set comprising first suggestions, each first suggestion comprising a respective first name and first resource identifier, wherein first features correspond to resources identified by the first resource identifiers;providing the query string to a local search provider executing on the computing device, and receiving therefrom a second suggestion set, the second suggestion set comprising second suggestions, each second suggestion comprising a respective second name and second resource identifier, wherein second features correspond to resources identified by the second resource identifiers;obtaining first rankings for the first suggestions, the first rankings computed according to the first features;obtaining second rankings for the second suggestions, the second rankings computed according to the second features;obtaining third rankings for the first and second suggestion sets, respectively;displaying a first representation for the first suggestion set and a second representation for the second suggestion set, wherein the first and second representations are displayed relative to each other according to the third rankings, wherein the first representation comprises first names displayed adjacent to each other and in order according to the first rankings, wherein the second representation comprises second names displayed adjacent to each other and in order according to the rankings of the second suggestions, and wherein the displayed first names and second names are user-selectable; andresponding to a user input selecting one of the displayed first or second names by requesting the first or second resource corresponding thereto.

17. Computer-readable storage hardware according to claim 16, wherein the resource identifier associated with the selected first or second name is used to request the first or second resource.

18. Computer-readable storage hardware according to claim 16, wherein the first, second, and third rankings are computed by the computing device.

19. Computer-readable storage hardware according to claim 16, wherein the first representation comprises instances of a first icon respectively graphically associated with the displayed first names, and wherein the second representation comprises instances of a second icon respectively graphically associated with the displayed second names.

20. Computer-readable storage hardware according to claim 16, wherein the local search provider comprises a command history, a file system, settings.