Method and apparatus for improving querying

• US 20040177061A1
• Filed: 03/05/2003
• Published: 09/09/2004
• Est. Priority Date: 03/05/2003
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Xu, Zhichen, Mahalingam, Mallik, Tang, Chunqiang
• Granted Patent: US 7,043,470 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/1834 implemented based on peer-to-peer networks, e.g. gnutella p2p communication protocols H04L67/104
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: H04L 67/104 Peer-to-peer [P2P] networks
• CPC Class Codes: H04L 67/1059 Inter-group management mechanisms, e.g. splitting, merging or interconnection of groups
• CPC Class Codes: H04L 67/1065 Discovery involving distributed pre-established resource-based relationships among peers, e.g. based on distributed hash tables [DHT] pre-configuration of logical or physical connections H04L67/1053
• CPC Class Codes: H04L 67/1093 Some peer nodes performing special functions
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

In a method of improving querying, a plurality of items of information are transformed into a plurality of vectors, where each item of information of the plurality of items of information is mapped to a respective vector of said plurality of vectors. A plurality of vectors is associated with a subset of nodes selected from a plurality of nodes. Each vector associated with a respective node of said plurality of nodes is based on a similarity between each vector of said plurality of vectors.

Claims

1. A method of improving querying, said method comprising:

transforming a plurality of items of information into a plurality of vectors, each item of information of said plurality of items of information being mapped to a respective vector of said plurality of vectors; and

associating said plurality of vectors to a subset of nodes selected from a plurality of nodes, each vector associated with a respective node of said plurality of nodes based on a similarity between each vector of said plurality of vectors.

2. The method according to claim 1, further comprising:

transforming said plurality of items of information into said plurality of vectors wherein, each item of information of said plurality of items of information is a document, video, or audio file.

3. The method according to claim 1, further comprising:

forming an auxiliary overlay network from said subset of nodes;

forming a plurality of zones; and

associating each zone of said plurality of zones with a respective node of said subset of nodes.

4. The method according to claim 1, wherein said transformation further comprises:

transforming said plurality of items of information into said plurality of vectors based on a vector space modeling.

5. The method according to claim 1, wherein said transformation further comprises:

transforming said plurality of items of information into said plurality of vectors based on a latent semantic indexing.

6. The method according to claim 1, further comprising:

receiving a query for requested information; and

transforming said query into a vectorized query.

7. The method according to claim 6, wherein said vectorized query is based on one of vector space modeling and a latent semantic indexing.

8. The method of according to claim 6, further comprising:

forming an auxiliary overlay network from said subset of nodes; and

locating said requested information based on routing said vectorized query through said auxiliary overlay network.

9. The method according to claim 1, further comprising:

forming groups of similar vectors from said plurality of vectors; and

associating each group of similar vectors of said groups of similar vectors with a respective node of said subset of nodes.

10. The method according to claim 9, further comprising:

ordering each group of similar vectors based on similarity between each group of similar vectors.

11. An apparatus for of improving querying, said apparatus comprising:

means for transforming a plurality of items of information into a plurality of vectors, each item of information of said plurality of items of information being mapped to a respective vector of said plurality of vectors; and

means for associating said plurality of vectors to a subset of nodes selected from a plurality of nodes, each vector associated with a respective node of said plurality of nodes based on a similarity between each vector of said plurality of vectors.

12. The apparatus according to claim 11, further comprising:

means for transforming said plurality of items of information into said plurality of vectors wherein, each item of information of said plurality of items of information is a document, video, or audio file.

13. The apparatus according to claim 11, further comprising:

means for forming an auxiliary overlay network from said subset of nodes;

means for forming a plurality of zones; and

means for associating each zone of said plurality of zones with a respective node of said subset of nodes.

14. The apparatus according to claim 11, wherein said means for transformation further comprises means for transforming said plurality of items of information into said plurality of vectors based on a vector space modeling.

15. The apparatus according to claim 11, wherein said means for transformation further comprises means for transforming said plurality of items of information into said plurality of vectors based on a latent semantic indexing.

16. The apparatus according to claim 11, further comprising:

means for receiving a query for requested information; and

means for transforming said query into a vectorized query.

17. The apparatus according to claim 16, wherein said vectorized query is based on one of vector space modeling and a latent semantic indexing.

18. The apparatus of according to claim 16, further comprising:

means for forming an auxiliary overlay network from said subset of nodes; and

means for locating said requested information based on routing said vectorized query through said auxiliary overlay network.

19. The apparatus according to claim 11, further comprising:

means for forming groups of similar vectors from said plurality of vectors; and

means for associating each group of similar vectors of said groups of similar vectors with a respective node of said subset of nodes.

20. The apparatus according to claim 19, further comprising:

means for ordering each group of similar vectors based on similarity between each group of similar vectors.

21. A system for querying, said system comprising:

a plurality of peers forming a peer-to-peer network;

an auxiliary overlay network configured to be implemented by a subset of peers of said plurality of peers; and

a peer search module configured to be executed by each peer of said subset of peers and configured to implement said auxiliary overlay network, wherein said peer search module is further configured to store key pairs configured to comprise of a vector and a pointer to a location of information.

22. The system according to claim 21, wherein said peer search module is further configured to store key pairs based on similarities of respective vectors of said key pairs.

23. The system according to claim 21, wherein said vector is generated from a selected item of information based on one of a vector space modeling and a latent semantic indexing.

24. The system according to claim 21, wherein said peer search module is further configured to receive a query for a selected item of information and to convert said query into a vectorized query.

25. The system according to claim 24, wherein said peer search module is further configured to route said vectorized query through said auxiliary overlay network to search for said selected information.

26. The system according to claim 25, wherein said vectorized query is based at least on one of said vector space modeling and said latent semantic indexing.