Trending Topic Identification From Social Communications

• US 20140358885A1
• Filed: 05/29/2013
• Published: 12/04/2014
• Est. Priority Date: 05/29/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Sun, Lei, Zhou, Enkai, Haoran, Li, Sun, Hoachuan
• Granted Patent: US 9,275,156 B2
• Time in Patent Office: Days
• US Class Current: 707/708
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing
• CPC Class Codes: G06F 16/9558 Details of hyperlinks; Management of linked annotations
• CPC Class Codes: G06Q 10/101 Collaborative creation, e.g. joint development of products or services
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Systems and methods for identifying trending topics from social communications are presented. Social communications from one or more social networking sites are obtained. Links within the social communications are extracted. In various embodiments, the extracted links are filtered such that the linked content corresponding to the extracted links that can be categorized as news are retained. Topic descriptors for the retained links are generated and the topic descriptors are stored in a trending topic data store as trending topics. According to various embodiments, the topic descriptors are stored in the trending topic data store with the extracted links.

Claims

1. A computer-implemented method for identifying trending topics on social networking services, the method comprising:

obtaining a plurality of social communications posted on one or more social networking services;

extracting a link to content from at least one of the plurality of social communications;

determining a set of potential topic descriptors for the linked content;

selecting a topic descriptor from the set of potential topic descriptors; and

storing the selected topic descriptor as a trending topic.

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

extracting a plurality of links from the plurality of social communications, each of the plurality of links corresponding to a different content; and

for each of the plurality of links;

categorizing the linked content according to news or non-news; and

retaining the links to linked content categorized as news.

3. The computer-implemented method of claim 2 further comprising:

for each of the plurality of links;

determining a set of potential topic descriptors for the linked content;

selecting a topic descriptor from the set of potential topic descriptors; and

storing the selected topic descriptor as a trending topic.

4. The computer-implemented method of claim 3, wherein determining a set of potential topic descriptors for the linked content comprises executing a lexical analysis tool to determine the set of potential descriptors, wherein the lexical analysis tool is configured to:

extract key terms and phrases from the linked content; and

generate a set of potential topic descriptors from the extracted key terms and phrases.

5. The computer-implemented method of claim 4, wherein determining a set of potential topic descriptors for the linked content further comprises:

comparing the link with one or more search logs obtained from at least one search engine;

identifying search queries from the search logs that resulted in search results that included the link; and

including the identified search queries as potential topic descriptors in the set of potential topic descriptors.

6. The computer-implemented method of claim 5, wherein selecting a topic descriptor from the set of potential topic descriptors comprises:

scoring each potential topic descriptor of the set of potential topic descriptors; and

selecting the highest scoring potential topic descriptor as the selected topic descriptor.

7. The computer-implemented method of claim 6, wherein scoring each potential topic descriptor of the set of potential topic descriptors comprises scoring each potential topic descriptor according to an estimation of how informative is the potential topic descriptor using a lexical analysis tool.

8. The computer-implemented method of claim 7, wherein scoring each potential topic descriptor of the set of potential topic descriptors further comprises, for each potential topic descriptor that was identified as a search query that resulted in search results that included a corresponding link, further scoring the potential topic descriptor according to the frequency of the search query in the search logs.

9. The computer-implemented method of claim 3, wherein determining a set of potential topic descriptors for the linked content comprises:

comparing the link with one or more search logs obtained from at least one search engine;

identifying search queries from the search logs that resulted in search results that included the link; and

including the identified search queries as potential topic descriptors in the set of potential topic descriptors.

10. The computer-implemented method of claim 9, wherein selecting a topic descriptor from the set of potential topic descriptors comprises:

scoring each potential topic descriptor of the set of potential topic descriptors; and

selecting the highest scoring potential topic descriptor as the selected topic descriptor.

11. The computer-implemented method of claim 10, wherein scoring each potential topic descriptor of the set of potential topic descriptors comprises, for each potential topic descriptor that was identified as a search query that resulted in search results that included a corresponding link, further scoring the potential topic descriptor according to the frequency of the search query in the search logs.

12. The computer-implemented method of claim 3, wherein selecting a topic descriptor from the set of potential topic descriptors comprises:

scoring each potential topic descriptor of the set of potential topic descriptors; and

selecting the highest scoring potential topic descriptor as the selected topic descriptor.

13. A computer-readable medium bearing computer-executable instructions which, when executed on a computing system comprising at least a processor executing instructions retrieved from the medium, carry out a method for identifying trending topics on social networking services, the method comprising:

obtaining a plurality of social communications posted on one or more social networking services;

extracting a plurality of links to content from the plurality of social communications, wherein each of the plurality of links is a link to different content; and

for each of the plurality of links to content;

categorizing the linked content according to news or non-news;

determining a set of potential topic descriptors for the linked content categorized as news;

selecting a topic descriptor from the set of potential topic descriptors; and

storing the selected topic descriptor as a trending topic.

14. The computer-readable medium of claim 13, wherein determining a set of potential topic descriptors for the linked content comprises:

executing a lexical analysis tool to determine the set of potential descriptors, and wherein the lexical analysis tool is configured to;

extract key terms and phrases from the linked content; and

generate a set of potential topic descriptors from the extracted key terms and phrases.

15. The computer-readable medium of claim 14, wherein determining a set of potential topic descriptors for the linked content further comprises:

comparing the link with one or more search logs obtained from at least one search engine;

identifying search queries from the search logs that resulted in search results that included the link; and

including the identified search queries as potential topic descriptors in the set of potential topic descriptors.

16. The computer-readable medium of claim 15, wherein selecting a topic descriptor from the set of potential topic descriptors comprises:

scoring each potential topic descriptor of the set of potential topic descriptors; and

selecting the highest scoring potential topic descriptor as the selected topic descriptor.

17. The computer-readable medium of claim 16, wherein scoring each potential topic descriptor of the set of potential topic descriptors comprises scoring each potential topic descriptor according to an estimation of how informative is the potential topic descriptor using a lexical analysis tool.

18. The computer-readable medium of claim 17, wherein scoring each potential topic descriptor of the set of potential topic descriptors further comprises, for each potential topic descriptor that was identified as a search query that resulted in search results that included a corresponding link, further scoring the potential topic descriptor according to the frequency of the search query in the search logs.

19. A computer system for identifying trending topics on social networking sites, the system comprising a processor and a memory, wherein the processor executes instructions stored in the memory as part of or in conjunction with additional components to identify trending topics on social networking services, the additional components comprising:

a social communication retrieval component to obtain social communications from one or more social networking sites;

a link extraction component configured to extract a plurality of links to content from the social communications, each of the plurality of links corresponding to a distinct item of content;

a news evaluation component configure to filter the plurality of links to content according to whether or not the linked content can be categorized as news, retaining those links to content that can be categorized as news; and

a topic descriptor generator that, for each link to content that can be categorized as news, generates a topic descriptor for each of the items of content and stores the generated topic descriptor with the corresponding link in a trending topic data store as a trending topic.

20. The computer system of claim 19, wherein for each link to content that can be categorized as news, the topic descriptor generator is configured to:

determine a first set of topic descriptors by;

extracting key terms and phrases from linked content; and

generating one or more potential topic descriptors from the extracted key terms and phrases;

determine a second set of topic descriptors by;

comparing a link with one or more search logs obtained from at least one search engine;

identifying search queries from the search logs that resulted in search results that included the link; and

including the identified search queries as potential topic descriptors in the set of potential topic descriptors;

combine the first and second set of topic descriptors;

score each topic descriptor of the combined set of topic descriptors; and

select the optimal scoring topic descriptor as the generated topic descriptor for the link.