TRACKING SIGNIFICANT TOPICS OF DISCOURSE IN FORUMS

• US 20100169327A1
• Filed: 12/31/2008
• Published: 07/01/2010
• Est. Priority Date: 12/31/2008
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Lindsay, Robert Taaffe, DiPersia, Blaise Andrew
• Granted Patent: US 9,521,013 B2
• Time in Patent Office: Days
• US Class Current: 707/750
• CPC Class Codes: G06F 16/3335 Syntactic pre-processing, e.g. stopword elimination, stemming
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/00 User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
• 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
• CPC Class Codes: H04L 67/306 User profiles

Abstract

Users in public forums often mention certain topics in the course of their discussions. Member'"'"'s comments in messages to other members are analyzed to obtain terms that co-occur with topics. Frequencies of co-occurrence of a term with topics are normalized based on frequency of the term in a random sample of message. The terms are ranked by their normalized frequency of co-occurrence with a topic in messages. The top terms are selected based on their rank. Analysis of demographic information associated with members that mentioned top terms associated with a topic is displayed in graphical format that highlights the relationship between the age, gender, and usage of the top terms over time. The demographic information presented includes average age of members that mentioned a top term or their gender information within a selected time interval.

Claims

1. A computer implemented method of selecting terms occurring in messages sent by members of a public forum to other members, wherein each term is associated with a topic, the method executed by a computer system, and comprising:

identifying a plurality of messages sent by members of the public forum, wherein each message in the plurality of messages contains the topic;

collecting a plurality of terms occurring in the plurality of messages;

ranking each term belonging to the plurality of terms, wherein the rank is determined based on criteria comprising frequency of occurrence of each term in the plurality of messages;

selecting a top terms set from the plurality of terms based on the rank of each term; and

storing the top terms in a computer readable storage medium.

2. The method of claim 1, wherein a term comprises a single word belonging to a message.

3. The method of claim 1, wherein a term comprises a sequence of multiple words belonging to a message.

4. The method of claim 1, wherein the criteria for determining the rank of each term further comprises:

a normalized frequency of occurrence of the term in the plurality of messages.

5. The method of claim 1, wherein the criteria for determining the rank of each term further comprises:

determining the ratio of the frequency of occurrence of the term in the plurality of messages to the frequency of occurrence of the term in a random sample of messages sent by members of the public forum.

6. The method of claim 1, further comprising:

determining a set dissimilarity score associated with a pair of terms; and

responsive to the dissimilarity score being less than a predetermined threshold, determining that each term belonging to the pair of terms is part of a phrase.

7. The method of claim 6, wherein determining a set dissimilarity score comprises determining a ratio of number of members that mentioned both the terms belonging to the pair of terms in at least one message to the number of members that mentioned at least one term belonging to the pair of terms in at least one message.

8. The method of claim 1 further comprising:

storing statistical information associated with a term belonging to the top terms in a computer readable storage medium.

9. The method of claim 8 wherein the statistical information comprises demographic information associated with a plurality of members, wherein each member belonging to the plurality of members sent at least a message containing the term belonging to the top terms.

10. A computer program product having a computer-readable storage medium storing computer-executable code for selecting terms occurring in messages sent by members of a public forum to other members, wherein each term is associated with a topic, the code comprising:

a top terms generator module configured to;

identify a plurality of messages sent by members of the public forum, wherein each message in the plurality of messages contains the topic;

collect a plurality of terms occurring in the plurality of messages;

rank each term belonging to the plurality of terms, wherein the rank is determined based on criteria comprising frequency of occurrence of each term in the plurality of messages; and

select a top terms set from the plurality of terms based on the rank of each term; and

a top terms analytic engine module configured to;

format a subset of top terms in the top terms set for display.

11. The computer program product of claim 10, wherein the criteria for determining the rank of each term further comprises:

a normalized frequency of occurrence of the term in the plurality of messages.

12. The computer program product of claim 10, wherein the criteria for determining the rank of each term further comprises:

the ratio of the frequency of occurrence of the term in the plurality of messages to the frequency of occurrence of the term in a random sample of messages sent by members of the public forum.

13. The computer program product of claim 10, wherein the top terms generator module is further configured to:

determine a set dissimilarity score associated with a pair of terms; and

responsive to the dissimilarity score being less than a predetermined threshold, determine that each term belonging to the pair of terms is part of a phrase.

14. The computer program product of claim 13, wherein determining a set dissimilarity score comprises determining a ratio of number of members that mentioned both the terms belonging to the pair of terms in at least one message to the number of members that mentioned at least one term belonging to the pair of terms in at least one message.

15. The computer program product of claim 10 wherein the top terms generator is further configured to:

store statistical information associated with a term belonging to the top terms in a computer readable storage medium.

16. The computer program product of claim 15 wherein the statistical information comprises demographic information associated with a plurality of members, wherein each member belonging to the plurality of members sent at least a message containing the term belonging to the top terms.