VISUALIZING CONFLICTS IN ONLINE MESSAGES

• US 20150106360A1
• Filed: 10/10/2013
• Published: 04/16/2015
• Est. Priority Date: 10/10/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Cao, Nan, Lin, Ching-Yung, Wang, Fei, Wen, Zhen
• Granted Patent: US 9,256,670 B2
• Time in Patent Office: Days
• US Class Current: 707/723
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/34 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Visualizing social media conflict is provided. Textual messages by a set of human users connected via a network regarding a particular topic are collected. Active users in the set of human users authoring a number of textual messages regarding the particular topic more than a threshold number of textual messages are selected. Keywords are selected that occur more than a threshold number of times within the textual messages regarding the particular topic. A sentiment score is computed for each of the keywords occurring more than the threshold number of times within the textual messages using a keyword co-occurrence graph. A sentiment of each of the active users is determined based on the computed sentiment score of each of the selected keywords that are authored by a particular active user.

Claims

1. A computer-implemented method for visualizing social media conflict, the method comprising:

collecting, by a computer, textual messages by a set of human users connected via a network regarding a particular topic;

selecting, by the computer, active users in the set of human users authoring a number of textual messages regarding the particular topic more than a threshold number of textual messages;

selecting, by the computer, keywords that occur more than a threshold number of times within the textual messages regarding the particular topic;

computing, by the computer, a sentiment score for each of the keywords occurring more than the threshold number of times within the textual messages using a keyword co-occurrence graph; and

determining, by the computer, a sentiment of each of the active users based on the computed sentiment score of each of the selected keywords that are authored by a particular active user.

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

selecting, by the computer, two distinct groups from the active users based on at least one of a relationship between the two distinct groups of active users and a determined degree of conflict between the two distinct groups of active users with regard to the particular topic.

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

determining, by the computer, a degree of sentiment of each of the two distinct groups of active users based on the determined sentiment of each of the active users within the two distinct groups of active users; and

computing, by the computer, a degree of polarity between the two distinct groups of active users based on the determined sentiment of each of the active users within each of the two distinct groups of active users.

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

generating, by the computer, a graph of the degree of sentiment of each of the two distinct groups of active users and the degree of polarity between the two distinct groups of active users as two smooth helical curves over a predetermined time interval, each of the two smooth helical curves overlapping at transition points representing sentiment overlap between the two distinct groups of active users and sentiment transition from one sentiment to another sentiment by the two distinct groups of active users.

5. The computer-implemented method of claim 4, further comprising:

connecting, by the computer, points on the two smooth helical curves with events containing keywords corresponding to sentiments of the active users at given times within the predetermined time interval.

6. The computer-implemented method of claim 5, further comprising:

displaying, by the computer, the generated graph of the two smooth helical curves with the connected points over the predetermined time interval; and

facilitating, by the computer, user interaction with the generated graph of the two smooth helical curves with the connected points.

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

ranking, by the computer, the keywords associated with the particular topic within time interval segments of online messages based on a number of occurrences of each keyword within a particular time interval segment to determine relevant keywords within the particular time interval segment, wherein the computer inserts the relevant keywords within the particular time interval segment into an event box located between sentiment streams of two different communities of social media users regarding the particular topic.

8. The computer-implemented method of claim 7, further comprising:

searching, by the computer, a sentiment dictionary for matching sentiments that correspond to keywords located in the keyword co-occurrence graph;

determining, by the computer, the keywords in the keyword co-occurrence graph that have a matching sentiment and using corresponding sentiment scores for the keywords having the matching sentiment;

propagating, by the computer, the corresponding sentiment scores of the keywords having the matching sentiment to adjacent keywords in the keyword co-occurrence graph that do not have a matching sentiment and corresponding sentiment score; and

computing, by the computer, the sentiment of each of the active users at each time interval segment of online messages based on sentiments scores of keywords in the keyword co-occurrence graph that were authored by a particular active user.

9. The computer-implemented method of claim 4, further comprising:

computing, by the computer, a first x-axis coordinate location on the graph for a first group of active users using a time value within the predetermined time interval;

computing, by the computer, a first y-axis coordinate location on the graph for the first group of active users using the determined sentiment of each active user within the first group of active users;

computing, by the computer, a second x-axis coordinate location on the graph for a second group of active users using the time value within the predetermined time interval;

computing, by the computer, a second y-axis coordinate location on the graph for the second group of active users using the determined sentiment of each active user within the second group of active users; and

generating, by the computer, the graph based on the computed first x-axis and first y-axis coordinate locations of the first group of active users and the computed second x-axis and second y-axis coordinate locations of the second group of active users.

10. A computer system for visualizing social media conflict, the data processing system comprising:

a bus system;

a storage device connected to the bus system, wherein the storage device stores computer readable program code; and

a processor unit connected to the bus system, wherein the processor unit executes the computer readable program code to collect textual messages by a set of human users connected via a network regarding a particular topic;

select active users in the set of human users authoring a number of textual messages regarding the particular topic more than a threshold number of textual messages;

select keywords that occur more than a threshold number of times within the textual messages regarding the particular topic;

compute a sentiment score for each of the keywords occurring more than the threshold number of times within the textual messages using a keyword co-occurrence graph; and

determine a sentiment of each of the active users based on the computed sentiment score of each of the selected keywords that are authored by a particular active user.

11. The computer system of claim 10, wherein the processor further executes the computer readable program code to select two distinct groups from the active users based on at least one of a relationship between the two distinct groups of active users and a determined degree of conflict between the two distinct groups of active users with regard to the particular topic.

12. A computer program product stored on a computer readable storage medium having computer readable program code encoded thereon that is executable by a computer for visualizing social media conflict, the computer program product comprising:

computer readable program code to collect textual messages by a set of human users connected via a network regarding a particular topic;

computer readable program code to select active users in the set of human users authoring a number of textual messages regarding the particular topic more than a threshold number of textual messages;

computer readable program code to select keywords that occur more than a threshold number of times within the textual messages regarding the particular topic;

computer readable program code to compute a sentiment score for each of the keywords occurring more than the threshold number of times within the textual messages using a keyword co-occurrence graph; and

computer readable program code to determine a sentiment of each of the active users based on the computed sentiment score of each of the selected keywords that are authored by a particular active user.

13. The computer program product of claim 12, further comprising:

computer readable program code to select two distinct groups from the active users based on at least one of a relationship between the two distinct groups of active users and a determined degree of conflict between the two distinct groups of active users with regard to the particular topic.

14. The computer program product of claim 13, further comprising:

computer readable program code to determine a degree of sentiment of each of the two distinct groups of active users based on the determined sentiment of each of the active users within the two distinct groups of active users; and

computer readable program code to compute a degree of polarity between the two distinct groups of active users based on the determined sentiment of each of the active users within each of the two distinct groups of active users.

15. The computer program product of claim 14, further comprising:

computer readable program code to generate a graph of the degree of sentiment of each of the two distinct groups of active users and the degree of polarity between the two distinct groups of active users as two smooth helical curves over a predetermined time interval, each of the two smooth helical curves overlapping at transition points representing sentiment overlap between the two distinct groups of active users and sentiment transition from one sentiment to another sentiment by the two distinct groups of active users.

16. The computer program product of claim 15, further comprising:

computer readable program code to connect points on the two smooth helical curves with events containing keywords corresponding to sentiments of the active users at given times within the predetermined time interval.

17. The computer program product of claim 16, further comprising:

computer readable program code to display the generated graph of the two smooth helical curves with the connected points over the predetermined time interval; and

computer readable program code to facilitate user interaction with the generated graph of the two smooth helical curves with the connected points.

18. The computer program product of claim 12, further comprising:

computer readable program code to rank the keywords associated with the particular topic within time interval segments of online messages based on a number of occurrences of each keyword within a particular time interval segment to determine relevant keywords within the particular time interval segment, wherein the computer inserts the relevant keywords within the particular time interval segment into an event box located between sentiment streams of two different communities of social media users regarding the particular topic.

19. The computer program product of claim 18, further comprising:

computer readable program code to search a sentiment dictionary for matching sentiments that correspond to keywords located in the keyword co-occurrence graph;

computer readable program code to determine the keywords in the keyword co-occurrence graph that have a matching sentiment and using corresponding sentiment scores for the keywords having the matching sentiment;

computer readable program code to propagate the corresponding sentiment scores of the keywords having the matching sentiment to adjacent keywords in the keyword co-occurrence graph that do not have a matching sentiment and corresponding sentiment score; and

computer readable program code to compute the sentiment of each of the active users at each time interval segment of online messages based on sentiments scores of keywords in the keyword co-occurrence graph that were authored by a particular active user.

20. The computer program product of claim 15, further comprising:

computer readable program code to compute a first x-axis coordinate location on the graph for a first group of active users using a time value within the predetermined time interval;

computer readable program code to compute a first y-axis coordinate location on the graph for the first group of active users using the determined sentiment of each active user within the first group of active users;

computer readable program code to compute a second x-axis coordinate location on the graph for a second group of active users using the time value within the predetermined time interval;

computer readable program code to compute a second y-axis coordinate location on the graph for the second group of active users using the determined sentiment of each active user within the second group of active users; and

computer readable program code to generate the graph based on the computed first x-axis and first y-axis coordinate locations of the first group of active users and the computed second x-axis and second y-axis coordinate locations of the second group of active users.