Patent Application
20180330720
Chat bots provide a way for users to ask questions using natural language. A search engine provider may add chat bots developed by the search engine provider. These chat bots may be integrated into search results to enhance the search experience. Using chat bots from third parties have complications as the search engine provider does not control third party chat bots. Insurances that a third-party chat hot is operating to the rules and/or policies of the search engine provider may be needed.
Chat bots may be useful to answer questions using natural language. In addition, the input into a chat bot may be used to augment a search experience, such as by updating search results, suggesting additional searches, etc. A search engine provider may add chat bots developed by the search engine provider to enhance the search experience. In addition, an external company may have produced its own chat bot which may also be integrated into search results. These chat bots may be selected and integrated into a search results page. Within that context the user may engage with the chat bot, which may be a third-party chat bot, without navigating away from the search engine results page.
Chat bots may enhance the search results page with or without user interaction. For example, search results may be refreshed after a user clicks on a given query or link from within a chat bot. In another example, search engine results may be refreshed automatically based on the context of the conversation within the chat bot. These refreshes may happen automatically in the background.
Using third-party chat bots introduces some risk if a chat bot behaves unpredictably in a real-world environment of mixed input information, especially with the ability of self-learning from the conversation. Limiting what information users may send to a chat bot may limit the performance of the chat bot. The behavior of the chat bot may be monitored to prevent a chat bot from behaving against some set of rules. As part of registration or in an ongoing manner, a chat bot may be sent a set of queries/input and the response from the chat bot may be validated. For example, typical queries may be sent to the chat bot and the responses sent to one or more classifiers. The classifiers may determine if a response is adult nature, racy, offensive, etc. In addition, another set of queries may be sent that include adult language, racy language, etc., and those responses may be sent to the classifiers. The queries may also be sent to the classier. In an example, the adult language, racy, etc. portions of the query may be stripped or removed before providing to the classifier. In an example, the classifier may remove the query words or a subset of the query words from the responses before classifying the response. For example, a curse word may be included in a query and repeated in the response. The curse word may be filtered from both the query and the response before providing to the classifier. If any chat bot is determined to be a “bad bot” based upon the returned classifier data, the chat bot may be blocked from being returned as part of the search engine results page.
Classifiers may be employed post-registration as well. For example, classifiers may be used to monitor currently running chat bots. The same queries used for testing chat bots during registration may be used. In an example, responses from real time queries may be used. The monitoring may happen periodically. For example, the monitoring may occur at a predefined time or after certain number of launches of a chat bot. The table below summarizes some classifiers and how they may be implemented. For example, some classifiers may be implemented using machine learned classifier models, javascript, etc.
In an example, user queries are matched against the tags to determine if the chat bot should be returned in the search engine results. The registration information may also include a chat bot category. A chat bot may be provided by and registered by third party entities, that are not associated with the operator of the search platform. In addition, the chat bot may be integrated into search results provided by a search engine while the chat bot hosted and use data services at a third-party location. In an example, a group of chat bots that belong to the same category may also be returned. For example, a search for “travel bots” may return multiple chat bots that are associated with the travel category of chat bots.
The registration information and the chat bot may be passed through a bot framework service bus 110. The bot framework service bus 110 may verify the registration information. In an example, a bot review 170 service may be used to verify the registration information. As described above, classifier may be used to review chat bot results. These classifiers may be used by the bot review 170 service. A web channel service bus 120 may then receive the registration information and determine where to store the registration data. The bot framework service bus 110, the web channel service bus 120, or the bot review 170 service may verify the chat bot. For example, predefined queries may be sent to the chat bot and the responses to these queries collected. The collected responses may be sent to one or more classifiers. The classifiers determine if the responses match an area that may disqualify the chat bot. For example, classifiers for adult material, racy, offensive, malware, phishing, spam, and is the bot alive may be used. In an example, the predefined queries may be a list of initial questions to chat bots. For example, “hello”; “help”; “terms of service”; etc. may be sent to the chat bot and the responses collected and classified. In an example, the predefined queries may contain material that would be classified into one of the areas. For example, queries could be racy, offensive, or adult nature. These responses could also be passed through the classifiers. In an example, the queries may be used to remove words from the chat bot responses before being passed to the classifiers.
If the responses from the chat bot are classified in an inappropriate area, the chat bot may be disqualified from registering with the web application 160. The owner of the chat bot may be contacted with information, such as the queries and responses, indicating that the chat bot failed a classifier.
Chat bots may be stored in a hot storage 130. The registration information or parts of the registration information may be stored in the hot storage 130 or a web application storage 140. For example, the tags associated with a chat bot may be stored in the web application storage.
In an example, the web application 160 is a search engine. The search engine receives a query and may interact with a Silverlight application package (XAP) service 150 to process the query. The XAP service 150 may interact with the web application storage 140 to determine the search results for the query. In an example, other content frameworks may be used. The XAP service 150 may determine that one or more chat bots should be included in the search results. For example, the search query may be used to match against the chat hot tags that were supplied as part of the chat hot registration. In an example, the chat bot that is included in the search results is a third-party chat bot that was successfully registered, including passing through the various classifiers. Chat bots may include text and/or voice input and outputs.
In an example, when a chat bot is matched to be returned in the search results, the XAP service 150 may determine if the chat bot is launchable from within the search results. The registration information may provide an indication whether a chat bot is launchable. If the chat bot is launchable, the chat bot may be included within the search results. In an example, the chat bot may be provided within a rich answer. In an example, to be included within a rich answer, the search query may have a strong intent for the results to include chat bots. For example, a search query for “travel bots” may signal a strong intent for chat bots based upon the use of the term “bots” within the search query. The rich answer may include information on a chat bot such as a description of the chat bot, an icon, a chat button, and an add chat hot button to a user's bot list. When the chat button is selected, the chat bot may be launched. In an example, the chat bot is a third-party chat bot that is launched within the search engine without the user navigating from the search engine results page.
In an example, a chat bot may be attached to search results. For example, the uniform resource locator (URL) of search results may be used to find relevant chat bots. During the chat bot registration, the chat bot may be associated with a URL. The chat bot's associated URL may be matched with the URL of the search results. For example, the entire URL, the domain of the URL, or a portion of the domain may be matched against the search results. In an example, only the first n search results are matched against chat bot URLs. If a match is found, the chat bot may be attached to the search result. In an example, if a match is found the chat bot may be included in a rich answer within the search results. This process allows for chat bots to be returned as part of organic search results, which helps users discover available chat bots.
The search results may be augmented based upon the chat bot. For example, a chat bot may receive an indication that a link within the chat bot was clicked. The chat bot may provide this indication to the web application. The XAP service may use this indication to refine a user's initial query and provide updated search results. In another example, the indication may be used to create a new search whose results replace or are added to the original search results.
In an example, when a chat bot is placed within search results, the search engine results page may register events to listen to the chat bot's activity. When an event is received from a chat bot, the search engine results page may provide that event to the web application. Example events include, but are not limited to, a user clicking on a chat bot element that contains an action to interact with the search results; trigger words within the conversation of the chat bot; etc. The trigger words may originate with the user's input or from the chat bots output. Some events do not require a user's action to intend to update the search results. The update may occur without the user's intent such that the search results page is updated automatically from the user's perspective.
The web application may update the search results with results from a brand new query; may replace/update a portion of the results on the current search engine results page; and/or refresh a portion of the search engine results page such as a carousel.
The chat bot server 416 may respond with a response 436. The response may be rendered 438 in the chat bot window 414. In an example, the response may include a hyperlink to additional material. For example, a user interacting with an encyclopedia chat bot may be provided with a link to additional information regarding a subject. The browser 410 may receive a click on the link and provide the click 440 to the chat bot window 414. Using the registered callbacks, the chat bot window 414 may provide an indication that a link has been clicked 442 to the SERP 412. The SERF 412 may then send an indication of the clicked link to a search engine. The search engine may process the clicked link and determine the initial search results should be updated. The search engine may send updated search results to the SERP 412 which may then refresh the currently displayed search results 444.
As an example, a user may initially do a search asking for airfare to a particular location. In response, the search results may include links to travel sites as well as a link to a travel chat bot. Clicking the link to the travel chat bot may launch the travel bot within the SERP 412. The travel chat bot may be passed the user's initial query and initiate conversation by confirming the user is searching for airfare from the user's location to the particular location. In an example, the user's location may be based upon user configuration data that may also be passed to the travel chat bot when the conversation is initialized. The particular location may be parsed from the user's search query.
The user may confirm both the destination and the originating location. The travel chat bot may confirm the requested dates of travel. Upon receiving this information, the travel chat bot may provide a response that includes prices for the airfare for the user. In addition, the response may also include a link to available hotels. When this link is clicked, the chat bot window 414 may receive an indication that the hotel link has been selected. The chat bot window 414 may provide an indication to the SERP 412 that the hotel link was selected. In an example, the payload passed to the SERP 412 as part of the call back may include the selected link, relevant dates, and search keywords. The SERP 412 may use the provided keywords to do a new query with the search engine. The search engine may provide results for the hotel keyword search. These results may replace the original search results within the SERP 412. In an example, the hotel keyword search results may supplement the original search results without fully replacing the original search results. For example, the top three original search results may be kept and the hotel keyword search results may replace lower ranking original search results.
In another example, a message may be sent from the SERP 412 to the chat bot window 414 using the registered callbacks. Continuing with the above example, the SERP 412 may include results of various airlines. The chat bot window 414 may include a listing of various airfares from different airlines. If the user selects a specific airline in the SERP 412, the chat bot window 414 may be provided with the selected link, the title of the corresponding webpage, etc. The chat bot window 414 may pass this information to the chat bot server 416. The chat bot server 416 may parse the received data and determine the user is interested in a particular airline. The chat bot server 416 may then provide a new response to the chat bot window 414 that provides a listing of airfares for only the airline associated with the selected link. In another example, the chat bot server 416 may determine other airlines have lower airfare and provide a response that highlights the lower airfare on different airlines.
When the link 744 is clicked, the search engine results page may be updated. In
Computing device 800 may include a hardware processor 802 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), a main memory 804 and a static memory 806, some or all of which may communicate with each other via an interlink (e.g., bus) 808. The computing device 800 may further include a display unit 810, an input device 812 (e.g., a keyboard), and a user interface (UI) navigation device 814 (e.g., a mouse). In an example, the display unit 810, input device 812, and UI navigation device 814 may be a touch screen display. In an example, the input device 812 may include a touchscreen, a microphone, a camera (e.g., a panoramic or high-resolution camera), physical keyboard, trackball, or other input devices.
The computing device 800 may additionally include a storage device (e.g., drive unit) 816, a signal generation device 818 (e.g., a speaker, a projection device, or any other type of information output device), a network interface device 820, and one or more sensors 821, such as a global positioning system (GPS) sensor, compass, accelerometer, motion detector, or other sensor. The computing device 800 may include an input/output controller 828, such as a serial (e.g., universal serial bus (USB), parallel, or other wired or wireless (e.g., infrared (IR), near field communication (NFC), etc.) connection to communicate or control one or more peripheral devices (e.g., a printer, card reader, etc.) via one or more input/output ports.
The storage device 816 may include a computing-readable (or machine-readable) storage media 822, on which is stored one or more sets of data structures or instructions 824 (e.g., software) embodying or utilized by any one or more of the techniques or functions described herein. In an example, at least a portion of the software may include an operating system and/or one or more applications (or apps) implementing one or more of the functionalities described herein. The instructions 824 may also reside, completely or at least partially, within the main memory 804, within the static memory 806, and/or within the hardware processor 802 during execution thereof by the computing device 800. In an example, one or any combination of the hardware processor 802, the main memory 804, the static memory 806, or the storage device 816 may constitute computing device (or machine) readable media.
While the computer-readable storage media 822 is illustrated as a single medium, a “computer-readable storage media” or “machine-readable storage media” may include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) configured to store the one or more instructions 824.
In an example, a computer-readable storage media or machine-readable storage media may include any medium that is capable of storing, encoding, or carrying instructions for execution by the computing device 800 and that cause the computing device 800 to perform any one or more of the techniques of the present disclosure, or that is capable of storing, encoding or carrying data structures used by or associated with such instructions. Non-limiting computer-readable storage media examples may include solid-state memories, and optical and magnetic media. Specific examples of computer-readable storage media may include: non-volatile memory, such as semiconductor memory devices (e.g., Electrically Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM)) and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; Random Access Memory (RAM); and optical media disks. The computer-readable storage media is non-transitory in that the storage media does not consist of transitory propagating signals.
The instructions 824 may further be transmitted or received over a communications network 826 using a transmission medium via the network interface device 820 utilizing any one of a number of transfer protocols (e.g., frame relay, internet protocol (IP), transmission control protocol (TCP), user datagram protocol (UDP), hypertext transfer protocol (HTTP), etc.). Example communication networks may include a local area network (LAN), a wide area network (WAN), a packet data network (e.g., the Internet), mobile telephone networks (e.g., cellular networks), Plain Old Telephone (POTS) networks, and wireless data networks (e.g., Institute of Electrical and Electronics Engineers (IEEE) 302.3 family of standards known as Wi-Fi®, IEEE 302.16 family of standards known as WiMax®), IEEE 302.15.4 family of standards, a Long Term Evolution (LTE) family of standards, a Universal Mobile Telecommunications System (UMTS) family of standards, peer-to-peer (P2P) networks, among others. The network interface device 820 may use the transfer protocols to transmit data using transitory propagating signals.
In an example, the network interface device 820 may include one or more physical jacks (e.g., Ethernet, coaxial, or phone jacks) or one or more antennas to connect to the communications network 826. In an example, the network interface device 820 may include one or more wireless modems, such as a Bluetooth modem, a Wi-Fi modem or one or more modems or transceivers operating under any of the communication standards mentioned herein. In an example, the network interface device 820 may include a plurality of antennas to wirelessly communicate using at least one of single-input multiple-output (SIMO), multiple-input multiple-output (MIMO), or multiple-input single-output (MISO) techniques. In some examples, the network interface device 820 may wirelessly communicate using Multiple User MIMO techniques. In an example, a transmission medium may include any intangible medium that is capable of storing, encoding or carrying instructions for execution by the computing device 800, and includes digital or analog communications signals or like communication media to facilitate communication of such software.
Any of the computer-executable instructions for implementing the disclosed techniques as well as any data created and used during implementation of the disclosed embodiments can be stored on one or more computer-readable storage media. The computer-executable instructions can be part of, for example, a dedicated software application or a software application that is accessed or downloaded via a web browser or other software application (such as a remote computing application). Such software can be executed, for example, on a single local computer (e.g., any suitable commercially available computer) or in a network environment (e.g., via the Internet, a wide-area network, a local-area network, a client-server network (such as a cloud computing network), or other such network) using one or more network computers.
The above description is intended to be illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with others. Other embodiments may be used, such as by one of ordinary skill in the art upon reviewing the above description. Further, the following claims are hereby incorporated into the Detailed Description, with a claim standing on its own as a separate embodiment. The scope of the embodiments disclosed herein is to be determined with reference to the appended claims, along with the fill scope of equivalents to which such claims are entitled.
This application claims priority to U.S. Provisional Patent Application Ser. No. 62/503,573 filed May 9, 2017, entitled “CHAT BOTS WITHIN SEARCH RESULTS,” which is incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 62503573 | May 2017 | US |
