Method and system of classifying online data

• US 8,255,402 B2
• Filed: 04/23/2009
• Issued: 08/28/2012
• Est. Priority Date: 04/23/2008
• Status: Active Grant

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Wang, Fang, Duman, Hakan, Thompson, Simon G, Nguyen, Duong
• Primary Examiner(s): Casanova, Jorge A
• Application Number:
  US12/988,714
• Publication Number:
  US 20110035377A1
• Time in Patent Office: 1,223 Days
• Field of Search: 707/737
• US Class Current: 707/737
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/95 Retrieval from the web

Abstract

A method of automatically analyzing online posts, such that they may then be responded appropriately. The method may comprise: extracting a list of keywords from each of the plurality of posts; generating one or more keyword clusters based on the keywords extracted from each of the plurality of posts and classifying new posts in accordance with the one or more keyword clusters.

Claims

1. A system for forming an online community comprising a processing unit, volatile and/or non-volatile memory and one or more data storage devices, the processing being configured to:

(a) extract a plurality of keywords from a plurality of posts;

(b) generate one or more keyword clusters based on the plurality of keywords extracted from the plurality of posts, each of the one or more keyword clusters comprising a plurality of keywords;

(c) classify new posts in accordance with the one or more keyword clusters,

(d) allocate a new post to one or more communities in accordance with the result of classifying in (c); and

(e) allocate a user to one or more communities in accordance with the classification of the posts they create;

wherein each community comprises a keyword cluster which comprises a plurality of core keywords and one or more node keywords, the core keywords having an association with each other which exceeds a first threshold value, the one or more node keywords having an association with each other which exceeds a second threshold value, wherein the first threshold value is greater than the second threshold value.

2. A system according to claim 1, the processing unit is further ocnfigured to:

(e) send a message to one or more service agents, the one or more service agents being associated with the community to which the post is allocated in (d).

3. A system according to claim 1, the processing unit is further configured to:

(e) send a message to one or more users, the one or more users being associated with the community to which the post is allocated in (d).

4. A system according to claim 1, wherein the extraction in (a) comprises determination of a term frequency-inverse document frequency weighting for one or more potential keywords.

5. A system according to claim 1, wherein processing unit is further configured to:

(e) form one or more keyword associations based on the list of keywords extracted in (a), said in (e) being carried out after the extraction (a) and before the generation of the one or more keyword clusters in (b).

6. A system according to claim 5, wherein the keyword associations are formed in accordance with the correlation of co-occurring keyword pairs.

7. A system according to claim 1, wherein a clique percolation model is applied to determine one or more keyword clusters.

8. A non-transitory computer readable storage medium comprising computer executable code which upon execution by a computer performs operation comprising:

extracting a plurality of keywords from a plurality of posts;

generating one or more keyword clusters based on the the plurality of keywords extracted from the plurality of posts, each of the one or more keyword clusters comprising a plurality of keywords;

classifying in new posts in accordance with the one or more keyword clusters;

allocating a new post to one or more communities in accordance with the result of said classifying new posts; and

allocating a user to one or more communities in accordance with the classification of the posts they create;

wherein each community comprises a keyword cluster which comprises a plurality of core keywords and one or more node keywords, the core keywords having an association with each other which exceeds a first threshold value, the one or more node keywords having an association with each he which exceeds a second threshold value, wherein the first threshold value is greater than the second threshold value.

9. The storage medium according to claim 8, the operation further comprises:

sending a message to one or more service agents, the one or more service agents being associated with the community to which the post is allocated in said allocating the new post.

10. The storage medium according to claim 8, the operation further comprises:

sending a message to one or more users, the one or more users being associated with the community to which the post is allocated in said allocating the new post.

11. The storage medium according to claim 8, wherein said extracting comprises determining a term frequency-inverse document frequency weighting for one or more potential keywords.

12. The storage medium according to claim 8, wherein the operation further comprises forming one or more keyword associations based on the list of keywords extracted in said extracting, said forming being performed after said extracting and before said generating.

13. The storage medium according to claim 12, wherein the keyword associations are formed in accordance with the correlation of co-occurring keyword pairs.

14. The storage medium according to claim 8, wherein a clique percolation model is applied to determine one or more keyword clusters.

15. A method comprising:

extracting, using a processing unit having at least one computer processor, a plurality of keywords from a plurality of posts;

generating one or more keyword clusters based on the plurality of keywords extracted from the plurality of posts, each of the one or more keyword clusters comprising a plurality of keywords;

classifying new posts in accordance with the one or more keyword clusters;

allocating a new post to one or more communities in accordance with the result of said classifying new posts; and

allocating a user to one more communities in accordance with the classification of the posts they create;

wherein each community comprises a keyword cluster which comprises a plurality of core keywords and one or more node keywords, the core keywords having an association with each other which exceeds a first threshold value, the one or more node keywords having an association with each other which exceeds a second threshold value, wherein the first threshold value is greater than the second threshold value.

16. The method according to claim 15, the method further comprises:

sending a message to one or more service agents or one or more users, the one or more service agents or the one or more users being associated with the community to which the post allocated in said allocating the new post.

17. The method according to claim 15, wherein said extracting comprises determining a term frequency-inverse document frequency weighting for one or more potential keywords.

18. The method according to claim 15, wherein the method further comprises forming one or more keyword associations based on the list of keywords extracted in said extracting, said forming being performed after said extracting and before said generating.

19. The method according to claim 18, wherein the keyword associations are formed in accordance with the correlation of co-occurring keyword pairs.

20. The method according to claim 15, wherein a clique percolation model is applied to determine one or more keyword clusters.