Implicit profile for use with recommendation engine and/or question router

• US 9,773,043 B2
• Filed: 08/03/2016
• Issued: 09/26/2017
• Est. Priority Date: 12/10/2008
• Status: Active Grant

• Current Assignee: Gartner, Inc.
• Original Assignee: Gartner, Inc.
• Inventors: Polonsky, Nir, Malangi, Girish
• Primary Examiner(s): Wong, Leslie
• Application Number:
  US15/227,420
• Publication Number:
  US 20160342660A1
• Time in Patent Office: 419 Days
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/1834 implemented based on peer-to-peer networks, e.g. gnutella p2p communication protocols H04L67/104
• CPC Class Codes: G06F 16/24 Querying
• CPC Class Codes: G06F 16/2457 with adaptation to user needs
• CPC Class Codes: G06F 16/252 between a Database Management System and a front-end application
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: Y10S 707/967 Peer-to-peer

Abstract

Methods and systems for creating an implicit profile for use by a recommendation engine or a question router is provided. User behavior on at least one of one or more electronic devices and an electronic communications network is tracked. User-related information relating to the user behavior is analyzed to extract or derive key words therefrom. The key words are stored in a profiles database as the implicit profile and used by the recommendation engine or question router to characterize user interests, expertise, and skills when matching a request from a querying user to a potential user or group of users having the relevant background to respond to the request.

Claims

1. An automated method for creating an implicit profile for use by at least one of a recommendation engine or a question router, comprising:

tracking user behavior on one or more electronic devices of a user and an electronic communications network used to access the recommendation engine or question router;

analyzing user-related information relating to the user behavior to extract and derive key words therefrom which are used to characterize user interests, expertise, and skills;

assigning key word weightings to each of the key words;

storing the key words and the key word weightings in a profiles database as the implicit profile;

wherein;

the key word weightings are assigned based on at least relevancy calculations of the user-related information and recency of the user-related information;

the relevancy calculations are based on at least a relation of the user-related information to current key words in the implicit profile;

at least the implicit profile is used by the recommendation engine or question router to provide expert recommendations or expert information to the user in response to a recommendation request or information request from the user;

access to the recommendation engine or question router is restricted and requires user login; and

the tracking is enabled via a crawling or searching application running on the one or more electronic devices of the user for searching various computer applications or storage locations on the one or more electronic devices for obtaining the user-related information.

2. The method in accordance with claim 1, wherein the user-related information comprises at least one of search terms used, documents read, documents opened, documents printed, documents saved, documents created, documents edited, documents commented on, annotations entered on documents, highlighted terms in documents, websites visited, webpages viewed, Internet searches conducted, ratings provided by the user on documents, products or services, user-created product or service reviews, multi-media items played, social forum threads opened, social forum threads participated in, people profiles opened, user items shared on the electronic communications network, shared items of others on the electronic communications network accessed by the user, user created content, online events, seminars, training courses, or webinars attended, in-person or online events, seminars, or training courses registered for, news or information feeds set up on the electronic communications network, emails written, emails received, blogs read, blog posts entered, software applications installed, computer hardware installed, software updates downloaded, and one or more of alerts, follows, and likes set up on the electronic communications network.

3. The method in accordance with claim 1, wherein:

the user-related information is obtained from user activity on a web site used to access the recommendation engine or the question router;

the user-related information comprises at least one of questions answered, answers provided, information and materials reviewed in answering questions, number of questions answered in a subject area, percentage of questions accepted for response, subject matter expertise based on answers submitted, number of times recommended as an expert in a subject area, user frequency of logging in to the web site, user preference for type of digital network, user preference for type of electronic device, user preferred information channels including one or more of reading or printing of documents, use of multimedia, and interacting socially in forums, and user open and click-through rates for relevant emails.

4. The method in accordance with claim 3, wherein the user-related information comprises at least one of user requests or user responses via the web site used to access the recommendation engine or the question router.

5. The method in accordance with claim 1, wherein the user-related information comprises user requests submitted to the recommendation engine or the question router via online forms, emails, or computer applications that are recorded in a request database to find at least one of experts, analysts, or peers, or to receive requested information or materials.

6. The method in accordance with claim 1, wherein the user-related information comprises user requests submitted to the recommendation engine or the question router or other linked web-based tools via online forms, emails, or computer applications that are recorded in a request database for at least one of vendor proposals, product demonstrations, and price quotes.

7. The method in accordance with claim 1, wherein the user-related information is obtained from electronic imprints from user interaction with the one or more electronic devices that are logged in a database.

8. The method in accordance with claim 7, wherein the one or more electronic devices comprise at least one of a computer, a tablet computer, a laptop, a smartphone, and an Internet enabled device.

9. The method in accordance with claim 7, wherein the electronic imprints are created by at least one of using a web interface, using a web browser, using a mobile application, using a computer application or program, sending or receiving an email, logging or recording a telephone call or voice message, manipulation of a self-reporting electronic system, and downloading or installing one or more of programs, applications, documents, multimedia content, music, and software updates.

10. The method in accordance with claim 1, wherein the key word weightings are further assigned based on one or more of an estimate of accuracy of the user-related information, type of the user-related information, source of the user-related information, amount of each type of the user-related information, time spent by user on each type or item of user-related information, and relation of the user-related information to information in the explicit profile of the user.

11. The method in accordance with claim 1, wherein the tracking comprises at least one of storing data from the user behavior at the time of the user behavior, searching the one or more electronic devices for data relating to the user behavior, and recording user interaction on the one or more electronic devices.

12. The method in accordance with claim 1, wherein the various computer applications comprise one or more of a word processing application, a web browser, an electronic calendar, an email program, spreadsheet applications, social media applications, messaging applications, and content editing, highlighting, and annotating programs.

13. The method in accordance with claim 1, wherein the storage locations comprise at least one of hard drive locations, file folders, document folders, web browser cookie folders, email folders, databases, spreadsheet folders, shared folders, networked folders, music folders, software application folders, media files, file directories, social directories, and activity logs.

14. The method in accordance with claim 1, wherein the key words are derived from the user-related information by applying at least one of lexical analysis, metadata analysis, and natural language processing analysis.

15. The method in accordance with claim 1, further comprising storing an explicit profile for the user in the profiles database together with the implicit profile for use by at least one of the recommendation engine and the question router, the explicit profile comprising profile data obtained by direct input from the user;

wherein a plurality of the implicit and explicit profiles are stored for each of a corresponding plurality of respective users in the profiles database.

16. The method in accordance with claim 15, wherein the implicit profile of the user is matched with the explicit profile of the user for use in processing the recommendation request or the information request.

17. The method in accordance with claim 16, wherein the key words, values, the key word weightings, and other information stored in the matching explicit and implicit profiles of the user are merged to create a merged profile for use by at least one of the recommendation engine and the question router.

18. The method in accordance with claim 15, wherein the question router automatically routes a question from a querying user to one or more of the other users on the electronic communications network based on a matching of key words obtained from the question with at least the implicit profiles of the one or more other users.

19. The method in accordance with claim 15, wherein:

the recommendation request comprises a request for an expert; and

the recommendation engine accepts the recommendation request for an expert from the user and recommends one or more of the other users on the electronic communications network as an expert based on a matching of key words obtained from the recommendation request with at least the implicit profiles of the one or more other users.

20. The method in accordance with claim 15, further comprising:

providing user feedback on relevancy of at least one of a recommendation request from the recommendation engine, a response to the recommendation request from a recommended peer, a question from the question router, and a response to the question to the user'"'"'s expertise, and

storing the feedback.

21. The method in accordance with claim 20, where the user'"'"'s feedback on relevancy is used to adjust the key word weightings in the implicit profile of at least one of the user or the recommended peer.

22. A system for creating an implicit profile for use by at least one of a recommendation engine or a question router, comprising:

an electronic communications network used to access the recommendation engine or question router;

one or more electronic devices for each user in communication with the electronic communications network;

a software application running on each of the electronic devices for each user for tracking user behavior on the one or more electronic devices and the electronic communications network used to access the recommendation engine or question router, the software application comprising a crawling or searching application running on each of the electronic devices for each user for searching various computer applications or storage locations on the electronic devices for obtaining user-related information relating to the user behavior;

one or more information databases for storing the user-related information;

an analyzer associated with the one or more databases and at least one of the recommendation engine or the question router for;

receiving and analyzing the user-related information relating to the user behavior;

extracting and deriving key words from the user-related information for use in characterizing user interests, expertise, and skills; and

assigning key word weightings to each of the key words; and

a profiles database associated with at least one of the recommendation engine or the question router for storing the key words and the key word weightings as the implicit profile;

wherein;

the key word weightings are assigned based on at least relevancy calculations of the user-related information and recency of the user-related information;

the relevancy calculations are based on at least a relation of the user-related information to current key words in the implicit profile; and

at least the implicit profile is used by the recommendation engine or question router to provide expert recommendations or expert information to the user in response to a recommendation request or information request from the user; and

access to the recommendation engine or question router is restricted and requires user login.

23. An automated method for creating an implicit profile for use by at least one of a recommendation engine or a question router, comprising:

tracking user behavior on one or more electronic devices of a user and on an electronic communications network used to access the recommendation engine or question router;

analyzing user-related information relating to the user behavior to extract or derive key words therefrom which are used to characterize user interests, expertise, and skills;

assigning key word weightings to each of the key words;

storing the key words and the key word weightings in a profiles database as the implicit profile;

determining whether an explicit profile has been provided by the user, the explicit profile comprising profile data obtained by direct input from the user;

in the event that an explicit profile has been provided by the user, merging the implicit profile and the explicit profile to create a merged profile for the user and storing the merged profile in the profiles database;

wherein;

the key word weightings are assigned based on at least relevancy calculations of the user-related information and recency of the user-related information;

in the event the explicit profile is not provided, the implicit profile is used by the recommendation engine or question router to provide recommendations or information to the user in response to a recommendation request or information request from the user; and

in the event the explicit profile is provided, the merged profile is used by the recommendation engine or question router to provide the recommendations or information to the user in response to a recommendation request or information request from the user.

24. The method in accordance with claim 23, wherein the tracking is enabled via a crawling or searching application running on the one or more electronic devices for searching various computer applications or storage locations on the one or more electronic devices for obtaining the user-related information.

25. A system for creating an implicit profile for use by at least one of a recommendation engine or a question router, comprising:

an electronic communications network used to access the recommendation engine or question router;

one or more electronic devices for each user in communication with the electronic communications network;

a software application running on each of the electronic devices for each user for tracking user behavior on at least one of the one or more electronic devices and the electronic communications network used to access the recommendation engine or question router;

one or more information databases for storing user-related information relating to the user behavior;

an analyzer associated with the one or more databases and at least one of the recommendation engine or the question router for;

receiving and analyzing the user-related information relating to the user behavior;

extracting or deriving key words from the user-related information for use in characterizing user interests, expertise, and skills; and

assigning key word weightings to each of the key words; and

a profiles database associated with at least one of the recommendation engine or the question router for storing the key words and the key word weightings as the implicit profile;

wherein;

it is determined whether an explicit profile has been provided by the user, the explicit profile comprising profile data obtained by direct input from the user;

in the event that an explicit profile has been provided by the user, merging the implicit profile and the explicit profile to create a merged profile for the user and storing the merged profile in the profiles database;

the key word weightings are assigned based on at least relevancy calculations of the user-related information and recency of the user-related information;

in the event the explicit profile is not provided, the implicit profile is used by the recommendation engine or question router to provide recommendations or information to the user in response to a recommendation request or information request from the user; and

in the event the explicit profile is provided, the merged profile is used by the recommendation engine or question router to provide the recommendations or information to the user in response to a recommendation request or information request from the user.