METHOD

• US 20110035377A1
• Filed: 04/23/2009
• Published: 02/10/2011
• 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, Thompson, Simon G, Duman, Hakan, Nguyen, Duong
• Granted Patent: US 8,255,402 B2
• Time in Patent Office: Days
• 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 analysing automatically analysing 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 method of analysing a plurality of online posts, the method comprising the steps of:

a) extracting a list of keywords from each of the plurality of posts;

b) generating one or more keyword clusters based on the keywords extracted from each of the plurality of posts and

c) classifying new posts in accordance with the one or more keyword clusters.

2. A method according to claim 1, the method comprising the further step of:

d) allocating a new post to a community in accordance with the result of step c).

3. A method according to claim 2, the method comprising the further step of:

e) 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 step d).

4. A method according to claim 2, the method comprising the further step of:

f) 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 step d).

5. A method according to claim 1, wherein step

a) comprises the determination of a term frequency-inverse document frequency weighting for one or more potential keywords.

6. A method according to claim 1, wherein the method comprises the further step of

i) forming one or more keyword associations based on the list of keywords extracted in step a), step i) being carried out after step a) and before step b).

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

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

9. A method according to claim 1, wherein in step c), a new post can be classified as belonging to a keyword cluster if a similarity function exceeds a predetermined threshold.

10. A method according to claim 1, wherein in step c) the similarity function is a cosine similarity function.

11. A tangible computer readable medium comprising computer executable code for performing a method according to claim 1.

12. An apparatus comprising a processing unit, volatile and/or non-volatile memory and one ore more data storage devices, the apparatus being configured, in use, to execute a method according to claim 1.