Analyzing distributed group discussions

• US 9,674,128 B1
• Filed: 06/03/2016
• Issued: 06/06/2017
• Est. Priority Date: 03/06/2013
• Status: Active Grant

• Current Assignee: Deloitte Development LLC
• Original Assignee: Blab Incorporated
• Inventors: Browning, Randolph, Snelling, David Aaron, Papp, Stefan Nicholas, Zimmermann, Bernhard G., Downs, Oliver B.
• Primary Examiner(s): NGUYEN, MINH CHAU
• Application Number:
  US15/173,402
• Time in Patent Office: 368 Days
• Field of Search: 709203-206
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/04 Real-time or near real-time messaging, e.g. instant messaging [IM]
• CPC Class Codes: H04L 51/216 Handling conversation history, e.g. grouping of messages in sessions or threads
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Techniques are described for analyzing user-supplied information, including in at least some situations to predict future aspects of additional related information that will be supplied by users. The user-supplied information that is analyzed may, for example, include distributed group discussions that involve numerous users and occur via user comments made to one or more social networking sites and/or other computer-accessible sites. The analysis of user-supplied information may, for example, include determining particular topics that are of interest for a specified category during one or more periods of time, quantifying an amount of user interest in particular topics and the category during the period of time, predicting future amounts of user interest in the particular topics and the category during one or more future period of times, and taking one or more further actions based on the predicted information.

Claims

1. A computer-implemented method comprising:

analyzing, by one or more configured computing systems, contents of a plurality of textual comments to identify topics that are mentioned in the contents, wherein the plurality of textual comments are supplied by human users from multiple geographical locations to multiple information services during a specified prior time period;

using, by the one or more configured computing systems, the contents of the plurality of textual comments to automatically determine a subset of the identified topics that are part of a specified content category for the specified prior time period, the using of the contents including;

generating, by the one or more configured computing systems, a plurality of comment groups based on the identified topics, each of the generated comment groups being associated with one of the identified topics and including one or more of the plurality of textual comments based on the contents of the one or more textual comments mentioning the associated identified topic;

identifying, by the one or more configured computing systems, a subset of the plurality of textual comments that are associated with the specified content category based on the contents of the textual comments of the identified subset, the textual comments of the identified subset being included in multiple of the generated comment groups;

determining, by the one or more configured computing systems, a subset of the multiple generated comment groups that correspond to the specified content category for the specified prior time period, the determining including excluding at least one first comment group whose included textual comments appear in the identified subset less than a determined minimum threshold, the determining further including excluding at least one second comment group whose included textual comments appear in the identified subset more than a determined maximum threshold; and

selecting, by the one or more configured computing systems and for each generated comment group in the subset of generated comment groups, the associated topic for the generated comment group to include in the subset of the identified topics that are part of the specified content category for the specified prior time period, wherein the subset of the identified topics includes multiple topics;

providing, by the one or more configured computing systems, indications of the multiple topics as representing the specified content category for the specified prior time period;

tracking, by the one or more configured computing systems and for at least one comment group in the subset, changes between the specified prior time period and one or more other time periods in at least one of textual comments supplied from the multiple geographical locations for the topic associated with the at least one comment group, or textual comments supplied to the multiple information sources for the topic associated with the at least one comment group; and

providing, by the one or more configured computing systems, information about the tracked changes between the specified prior time period and the one or more other time periods for the topic associated with the at least one comment group.

2. The method of claim 1 further comprising:

determining, by the one or more configured computing systems, and for one of the comment groups of the determined subset, a quantity of the textual comments included in the one comment group;

determining, by the one or more configured computing systems and based at least in part on the determined quantity, one or more types of information to make available corresponding to at least one of the specified content category and the identified topic associated with the one comment group; and

supplying, by the one or more configured computing systems, one or more additional textual comments that are available to the human users and that have contents corresponding to the determined one or more types of information.

3. The method of claim 2 wherein the plurality of textual comments are part of user posts to one or more social networking services by the human users, and wherein the supplying of the one or more additional textual comments includes supplying the additional textual comments to at least one of the social networking services as additional posts.

4. The method of claim 1 further comprising:

determining, by the one or more configured computing systems, and for each of the comment groups of the determined subset, a quantity of the textual comments included in the comment group;

predicting, by the one or more configured computing systems, and for each of the comment groups of the determined subset, an expected quantity of future textual comments associated with the identified topic for the comment group that will be supplied by human users during a future time period, the predicting being based in part on the determined quantity for the comment group; and

providing, by the one or more configured computing systems, an indication of the predicted expected quantity of future textual comments for one or more of the comment groups of the determined subset.

5. The method of claim 1 further comprising:

determining, by the one or more configured computing systems, and for each of the comment groups of the determined subset, a quantity of the textual comments included in the comment group;

predicting, by the one or more configured computing systems, an expected quantity of future textual comments associated with the specified content category that will be supplied by human users during a future time period, the predicting being based in part on a combination of the determined quantities for the comment groups of the determined subset; and

providing, by the one or more configured computing systems, an indication of the predicted expected quantity of future textual comments for the specified content category.

6. The method of claim 1 wherein the tracking of the changes between the specified prior time period and the one or more other time periods for the at least one comment groups includes tracking changes between the specified prior time period and the one or more other time periods in the textual comments supplied from the multiple geographical locations for the topic associated with the at least one comment group, and further includes:

determining, by the one or more configured computing systems, and for each of the at least one comment groups, distinct subsets of the textual comments included in the comment group that are associated with each of the multiple geographical locations for the specified prior time period;

determining, by the one or more configured computing systems, and for each of the multiple geographical locations, a quantity of the textual comments included in the subset associated with the geographical location for the specified prior time period; and

providing, by the one or more configured computing systems, information about the determined quantity for each of one or more of the multiple geographical locations for the specified prior time period.

7. The method of claim 1 wherein the plurality of textual comments are posts supplied to social networking services, and wherein the tracking of the changes between the specified prior time period and the one or more other time periods for the at least one comment groups further includes tracking changes between the specified prior time period and the one or more other time periods in textual comments supplied by multiple users of the plurality of users, by:

determining, by the one or more configured computing systems, and for each of the at least one comment groups, distinct subsets of the textual comments included in the comment group that are associated with each of the multiple human users for the specified prior time period;

determining, by the one or more configured computing systems, and for each of the multiple human users, a quantity of the textual comments included in the subset associated with the human user for the specified prior time period; and

providing, by the one or more configured computing systems, information about the determined quantity for each of one or more of the multiple human users for the specified prior time period.

8. The method of claim 1 wherein the tracking of the changes between the specified prior time period and the one or more other time periods for the at least one comment groups includes tracking changes between the specified prior time period and the one or more other time periods in the textual comments supplied to the multiple information sources for the topic associated with the at least one comment group, and further includes:

determining, by the one or more configured computing systems, and for each of the at least one comment groups, distinct subsets of the textual comments included in the one comment group that are associated with each of the multiple information sources for the specified prior time period;

determining, by the one or more configured computing systems, and for each of the multiple information sources, a quantity of the textual comments included in the subset associated with the information source for the specified prior time period; and

providing, by the one or more configured computing systems, information about the determined quantity for each of one or more of the multiple information sources for the specified prior time period.

9. The method of claim 1 wherein the one or more configured computing systems are part of a group discussion prediction service, wherein the specified content category is indicated by a client of the group discussion prediction service, and wherein the providing of the indications of the multiple topics includes providing information to the client in accordance with criteria specified by the client.

10. The method of claim 1 further comprising:

performing an influence analysis of the textual comments included in one or more comment groups of the determined subset to identify a first author user of the human users who has greater influence in affecting comments of other users than one or more other second author users of the human users; and

providing one or more indications of the first author user and the performed influence analysis.

11. The method of claim 1 further comprising:

determining, with respect to the textual comments included in one or more comment groups of the determined subset, an increase or a decrease in a positive or negative sentiment relative to multiple textual comments supplied during the one or more other time periods; and

providing one or more indications of the determined increase or decrease in the positive or negative sentiment.

12. The method of claim 1 wherein the identified plurality of topics are each a term or phrase.

13. The method of claim 1 wherein the identified plurality of topics are each a hashtag.

14. The method of claim 1 further comprising:

determining, for each of the one or more other time periods, a plurality of comment groups that correspond to the specified content category for the additional prior time period and that have corresponding topics, wherein the corresponding topics for the determined plurality of comment groups of the other time periods are distinct from the multiple topics determined to correspond to the specified content category for the specified prior time; and

providing indications of one or more topic changes for the specified content category between the specified prior time period and the one or more other time periods.

15. A non-transitory computer-readable medium having stored contents that cause one or more computing systems to perform a method, the method comprising:

analyzing, by the one or more computing systems, a plurality of user-supplied content items to identify a plurality of attributes that are each associated with one or more of the plurality of user-supplied content items;

generating, by the one or more computing systems, a plurality of comment groups based on the identified attributes, each of the generated comment groups being associated with one of the identified attributes and including one or more of the plurality of user-supplied content items that have the associated identified attribute;

identifying, by the one or more computing systems, a subset of the plurality of user-supplied content items that are associated with a specified content category based on contents of the user-supplied content items of the identified subset, the user-supplied content items of the identified subset being included in multiple of the generated comment groups;

determining, by the one or more computing systems, a subset of the multiple generated comment groups that correspond to the specified content category, the determining including excluding one or more comment groups each having included user-supplied content items that appear in the identified subset less than a determined minimum threshold or more than a determined maximum threshold; and

providing, by the one or more computing systems, information identifying one or more determined topics for the specified content category based on the associated identified attributes for one or more of the generated comment groups in the determined subset.

16. The non-transitory computer-readable medium of claim 15 wherein the plurality of content items are supplied by multiple human users to multiple data sources, wherein some content items of the plurality of content items each has one or more associated tags that are supplied by one or more users to describe that content item, wherein other content items of the plurality of content items each has one or more associated tags that are supplied by at least one of the data sources to describe that content item, and wherein the stored contents further cause the one or more computing systems to quantify, based at least in part on tags associated with the content items of the plurality, information of one or more types about at least one comment group of the generated comment groups.

17. The non-transitory computer-readable medium of claim 16 wherein the stored contents further cause the one or more computing systems to, before the analyzing of the plurality of user-supplied content items, obtain content items from the at least one comment data source that meet one or more specified criteria.

18. The non-transitory computer-readable medium of claim 16 wherein the stored contents further cause the one or more computing systems to generate an aggregation of information that includes user-supplied content items in the generated comment groups of the determined subset and is based at least in part on the quantified information, including using weighting to reflect one or more factors for the quantified information of the generated aggregation.

19. The non-transitory computer-readable medium of claim 16 wherein the stored contents further cause the one or more computing systems to select one of the generated comment groups for further analysis based at least in part on one or more of the data sources associated with the selected one generated comment group.

20. The non-transitory computer-readable medium of claim 16 wherein the stored contents further cause the one or more computing systems to analyze the generated comment groups based on one or more metrics, and wherein the determining of the subset is further based at least in part on the analyzing of the generated comment groups and includes using a specified minimum certainty threshold.

21. The non-transitory computer-readable medium of claim 15 wherein the plurality of content items are comments supplied by a plurality of human users to one or more social networking services, wherein the identified plurality of attributes include a plurality of terms included in contents of the plurality of content items, and wherein the one or more computing systems are part of a group discussion prediction system.

22. The non-transitory computer-readable medium of claim 15 wherein the stored contents include software instructions that, when executed, further cause the one or more computing systems to perform, for each of a plurality of distinct time periods, the analyzing and the generating and the identifying and the determining and the providing based on content items supplied by users during the time period, and wherein the providing of the indications of the one or more determined multiple topics includes identifying a change in one or topics that are associated with the specified content category between multiple time periods.

23. The non-transitory computer-readable medium of claim 15 wherein the plurality of attributes include at least one of multiple geographical locations from which content items are supplied and multiple information sources to which content items are supplied, wherein the plurality of content items are supplied by users during a time period, and wherein the stored contents include software instructions that, when executed, further cause the one or more computing systems to:

track, by the one or more computing systems and for at least one comment group in the subset of generated comment groups, changes between the time period and the one or more other time periods in at least one of content items supplied from the multiple geographical locations for the at least one comment group, or content items supplied to the multiple information sources for the at least one comment group; and

provide further information about the tracked changes between the time period and the one or more other time periods for the at least one comment group.

24. The non-transitory computer-readable medium of claim 15 wherein the plurality of attributes include users who supplied content items, and words included in contents of content items.

25. A system, comprising:

one or more hardware processors of one or more computing systems; and

one or more modules that, when executed by at least one of the one or more hardware processors, cause the at least one hardware processors to determine multiple topics associated with a specified content category based on information supplied by users, the determining of the multiple topics including;

analyzing a plurality of user-supplied content items to identify a plurality of attributes that are each associated with one or more of the plurality of user-supplied content items, including identifying topics associated with the plurality of user-supplied content items, and wherein the plurality of user-supplied content items are supplied by a plurality of users during a time period;

using the identified attributes of the plurality of user-supplied content items to automatically determine a subset of the identified topics that are part of the specified content category for the time period, the using of the identified attributes including;

generating a plurality of comment groups based on the identified attributes, each of the generated comment groups being associated with one of the identified attributes and including one or more of the plurality of user-supplied content items that have the associated identified attribute;

identifying a subset of the plurality of user-supplied content items that are associated with the specified content category based on the identified attributes of the user-supplied content items of the identified subset, the user-supplied content items of the identified subset being included in multiple of the generated comment groups;

determining a subset of the multiple generated comment groups that correspond to the specified content category, the determining including excluding one or more comment groups each having included user-supplied content items that appear in the identified subset less than a determined minimum threshold or more than a determined maximum threshold; and

selecting, for each generated comment group in the subset of generated comment groups, an associated topic for the generated comment group to include in the determined subset of the identified topics that are part of the specified content category for the time period;

providing indications of the identified topics of the determined subset as being associated with the specified content category for the time period;

tracking changes, between the time period and one or more other time periods in topics, in topics determined as being associated with the specified content category; and

providing information about the tracked changes between the time period and the one or more other time periods in the topics determined as being associated with the specified content category.

26. The system of claim 25 wherein the plurality of content items are comments supplied to one or more social networking services, wherein the identified plurality of attributes include a plurality of terms included in contents of the plurality of content items, and wherein the one or more modules includes software instructions of a group discussion prediction system that are stored in one or more memories of the system for execution by the at least one hardware processors.

27. The system of claim 25 wherein the determining of the multiple topics associated with the specified content category includes performing the analyzing of user-supplied content items and the using of the identified attributes for each of a plurality of distinct time periods based on content items supplied by users during the time period, the plurality of time periods including the one or more other time periods.

28. The system of claim 25 wherein the plurality of attributes include at least one of multiple geographical locations from which contents item are supplied and multiple information sources to which content items are supplied, and wherein the one or more modules include software instructions that, when executed, further cause the at least one hardware processor to:

track, for at least one comment group in the subset of generated comment groups, changes between the time period and the one or more other time periods in at least one of content items supplied from the multiple geographical locations for the at least one comment group, or content items supplied to the multiple information sources for the at least one comment group; and

provide further information about the tracked changes between the time period and the one or more other time periods for the at least one comment group.

29. The system of claim 25 wherein the one or more modules consists of one or more means for performing the determining of the multiple topics associated with the specified content category based on the information supplied by the users.

30. The system of claim 25 wherein the plurality of attributes include users who supplied content items, and words included in contents of content items.