Interest tracking using shared search queries and interactions

• US 8,438,165 B2
• Filed: 05/12/2011
• Issued: 05/07/2013
• Est. Priority Date: 05/12/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Ickman, Steven, Laird-McConnell, Thomas Matthew, Portnoy, William Lawrence, McConnell, Christopher, Cheng, Lili, Raskino, David Arthur, Melander, Gregory Scott
• Primary Examiner(s): MOSER, BRUCE M
• Application Number:
  US13/105,920
• Publication Number:
  US 20120290603A1
• Time in Patent Office: 726 Days
• Field of Search: 707/999.003, 707/999.102, 707/999.107, 707/713, 707/769
• US Class Current: 707/738
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

When a user enters an information retrieval query, topic feed is generated for publication or distribution to a set of followers or recipients who have elected to receive the topic feed from the user or based on the subject matter content of the query. The topic feed is distributed and includes the query. The user'"'"'s interests are tracked based on text in the user'"'"'s queries and other text in the topic feed that the user interacts with.

Claims

1. A computer-implemented method of identifying user interests, using a computer processor, comprising:

receiving text from a shared topic feed distributed to the computer processor, the shared topic feed comprising at least one search query entered by a user in a public stream and a link associated with the search query, and wherein the shared topic feed further includes search results returned for the query submitted by the user along with a link corresponding to each of the search results;

receiving an indication that the user has interacted with a portion of the received text, including receiving an indication that the user has submitted the query entered by the user for execution by a search engine;

receiving marked text marked with an explicit interest indication that the user is interested in a subject matter or a person identified in the marked text;

identifying content words in the portion of the received text;

removing the marked text from the content words;

performing a statistical analysis on the content words, with the marked text removed, to identify the user interests, wherein the user interests include a topic of interest that includes at least one of a subject matter of interest and a person of interest;

identifying the user interests based on the explicit interest indication and the statistical analysis; and

displaying an indication to the user indicative of the identified user interests.

2. The computer-implemented method of claim 1 wherein identifying content words in the portion of the received text comprises identifying content words in the search query submitted by the user.

3. The computer-implemented method of claim 1 wherein receiving an indication that the user has interacted with the portion of the received text comprises:

receiving an indication that the user has selected at least one of the search results.

4. The computer-implemented method of claim 3 wherein identifying content words in the portion of the received text comprises identifying content words in the selected search result.

5. The computer-implemented method of claim 1 wherein the search query in the shared topic feed comprises a query submitted by a third party, and wherein receiving an indication that the user has interacted with the portion of received text comprises:

receiving an indication that the user has actuated the link associated with the query.

6. The computer-implemented method of claim 1 wherein the shared topic feed comprises a plurality of different posts, and further comprising:

providing, in the topic feed, a user input mechanism by which the user can comment on the plurality of posts in the shared topic feed, and wherein receiving an indication that the user has interacted with the portion of received text comprises receiving an indication that the user has commented on one of the posts, such that, the portion of text comprises text in the post on which the user commented.

7. The computer-implemented method of claim 1 wherein the shared topic feed comprises a plurality of different posts, and further comprising:

providing, in the shared topic feed, a user input mechanism by which the user can like individual ones of the plurality of posts in the shared topic feed, and wherein receiving an indication that the user has interacted with the portion of received text comprises receiving an indication that the user has liked one of the posts, such that, the portion of text comprises text in the post which the user liked.

8. The computer-implemented method of claim 1 wherein the shared topic feed comprises a plurality of posts, each post including a user identifier identifying a user that contributed to the post, along with a link corresponding to the user identifier, and wherein receiving an indication that the user has interacted with the portion of received text comprises:

receiving an indication that the user has actuated the actuable link corresponding to the user identifier such that the portion of the received text comprises text in the user identifier.

9. The computer-implemented method of claim 1 wherein identifying content words comprises:

performing linguistic analysis on the portion of the received text to identify the content words.

10. The computer-implemented method of claim 1 wherein identifying the user interests based on the content words identified comprises:

performing the statistical analysis to identify a frequency of occurrence of the content words in text with which the user has interacted.

11. The computer-implemented method of claim 1 wherein identifying the user interests based on the content words identified comprises:

identifying the user interests from a list of predefined topics of interest based on the statistical analysis, wherein the predefined topics of interest include people of interest and subject matter areas of interest.

12. The computer-implemented method of claim 1 wherein identifying the user interests based on the content words identified comprises:

identifying the user interests and generating a new topic of interest, corresponding to the identified user interests, based on the statistical analysis, wherein the new topic of interest includes one of a subject matter area of interest or a person of interest.

13. The computer-implemented method of claim 1 wherein the user can specify sources of the topic feed by specifying subject matter of posts to be received and people from whom posts are to be received, and further comprising:

receiving an input from the user indicating that the user wishes to receive posts in the shared topic feed from people, and about subject matter, in the topic of interest; and

including, in subsequent shared topic feed posts distributed to the computer processor, posts from the people, and about the subject matter, in the topic of interest.

14. The computer-implemented method of claim 1 and further comprising:

including, in subsequent shared topic feed posts distributed to the computer processor, posts from the people, and about the subject matter, in the marked text.

15. A computer implementing public search system, comprising:

a topic feed generator that receives input queries and generates a topic feed including posts of the input queries;

a feed distributor that distributes the topic feed to recipients;

a user interface component that displays the topic feed and receives user inputs indicative of a user interacting with text in the topic feed;

an interest tracking component that identifies topics of interest based on the text that the user interacted with in the topic feed, the feed distributor including in the topic feed displayed at the user interface component, posts corresponding to the identified topics of interest, the interest tracking component including an explicit component that identifies the topics of interest in text that is explicitly marked as of interest, and an implicit component that performs linguistic and statistical analysis on remaining text, with the explicitly marked text removed, that the user interacted with in the topic feed to identify the topics of interest; and

a processor that is a functional component of the system and is activated by the topic feed generator and the interest tracking component to facilitate generating the topic feed and identifying topics of interest.