GENERATING USER-CUSTOMIZED SEARCH RESULTS AND BUILDING A SEMANTICS-ENHANCED SEARCH ENGINE

• US 20100268702A1
• Filed: 04/14/2010
• Published: 10/21/2010
• Est. Priority Date: 04/15/2009
• Status: Active Grant

• Current Assignee: Fiver LLC
• Original Assignee: EVRI Incorporated
• Inventors: Spivack, Nova, Wissner, James M.
• Granted Patent: US 9,037,567 B2
• Time in Patent Office: Days
• US Class Current: 707/711
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/955 using information identifiers, e.g. uniform resource locators [URL]

Abstract

Systems and methods for generating user-customized search results and building a semantics-enhanced search engine are disclosed. In one aspect, embodiments of the present disclosure include a method, which may be implemented on a system, of generating user-customized search results using user-defined semantic types. The method includes, identifying a first set of URI patterns that are associated with a first set of semantic types defined by a first user, storing the first set of URI patterns in a database embodied in a computer-readable storage medium, and/or semantically categorizing a first set of search results for the first user, as having content related to one or more of the first set of semantic types defined by the first user. The first set of search results can be categorized using the first set of URI patterns.

Claims

1. A method for generating user-customized search results using user-defined semantic types, the method, comprising:

identifying a first set of URI patterns that are associated with a first set of semantic types defined by a first user;

storing the first set of URI patterns in a database embodied in a computer-readable storage medium;

semantically categorizing a first set of search results for the first user, as having content related to one or more of the first set of semantic types defined by the first user;

wherein, the first set of search results are categorized using the first set of URI patterns.

2. The method of claim 1, further comprising:

identifying a second set of URI patterns that are associated with a second set of semantic types defined by a second user;

storing the second set of URI patterns in a database embodied in the computer-readable storage medium.

3. The method of claim 2, further comprising:

semantically categorizing a second set of search results for the second user, as having content related to one or more of the second set of semantic types defined by the second user;

wherein, the second set of search results are categorized using the second set of URI patterns.

4. The method of claim 3,

wherein, the first set of search results are identified in response to a keyword search requested by the first user;

wherein, the second set of search results are identified in response to the keyword search requested by the second user that is a same keyword search as the first user'"'"'s keyword search.

5. The method of claim 3, further comprising:

presenting the first set of search to be visually categorized to the first user as having content related to the one or more of the first set of semantic types;

presenting the second set of search results to be visually categorized to the second user as having content related to the one or more of the second set of semantic types.

6. The method of claim 1, further comprising, parsing a web-dialogue regarding a particular webpage in which the first user is participating according to a grammar.

7. The method of claim 6, further comprising, categorizing the particular web page, based on the parsing, as having content related to one or more of the first set of semantic types defined by the first user.

8. The method of claim 1,

wherein, the first set of URI patterns are identified from a first set of URIs of web pages having content specified by the first user as having content of the first set of semantic types;

and

wherein, the second set of URI patterns are identified from a second set of URIs of web pages having content specified by the second user as having content of the second set of semantic types.

9. The method of claim 1, wherein, the first set of search results are categorized by comparing associated URIs of each of the first set of search results with the first set of URI patterns.

10. The method of claim 1, further comprising:

identifying a first additional set of URI patterns that are associated with a first set of attributes of a semantic type that are defined by the first user;

semantically categorizing the first set of search results for the first user, as having content related to one or more of the first set of attributes of the semantic type;

wherein, the first set of search results are categorized using the first additional set of URI patterns.

11. The method of claim 10, wherein, the first additional set of URI patterns are identified from a first additional set of URIs of web pages having content specified by the first user as having content related to one or more of the first set of attributes of the semantic type.

12. A method of building a semantics-enhanced search engine, the method, comprising:

defining a location identifier pattern for identifying objects having content of a semantic type;

storing the location identifier pattern in a computer database embodied on a computer-readable storage medium;

analyzing a sitemap stored on a host server for a particular domain;

identifying, from the sitemap, multiple location identifiers of the particular domain matching the location identifier pattern defined for the semantic type;

indexing each of the multiple location identifiers as being associated with objects having content of the semantic type.

13. The method of claim 12,

wherein, the location identifier is a Universal Resource Identifier (URI) or Universal Resource Locator (URL);

wherein, the objects are web pages.

14. The method of claim 13, further comprising, associating the web pages with an ontology defining the semantic type.

15. The method of claim 14, further comprising, associating the URL or URI with the ontology.

16. The method of claim 13, further comprising, converting HTML or XHTML content in the web pages to semantic data for use in the semantics-enhanced search engine.

17. The method of claim 16, wherein, the HTML or XHTML content is converted using the GRDDL (Gleaning Resource Descriptions from Dialects of Languages) markup format.

18. The method of claim 12, further comprising:

defining a grammar for parsing web-based dialogue.

19. The method of claim 18, further comprising:

parsing a web-based dialogue regarding a particular object according to the grammar;

identifying the particular object as having content of or related the semantic type based on the parsing.

20. The method of claim 19, further comprising, detecting a predicate and object of the web-based dialogue using the grammar.

21. The method of claim 19, further comprising, generating RDF triplets from the web-based dialogue.

22. The method of claim 21, wherein, the RDF triples and the grammar are used for conversational search in the semantics-enhanced search engine.

23. The method of claim 18 wherein, the web-based dialogue is one or more of, online chat, a review, a comment, a status update, and a posting.

24. The method of claim 18, wherein, the grammar is a general grammar.

25. The method of claim 18, wherein, the grammar is a domain-specific grammar.

26. The method of claim 12, further comprising,

data mining content of objects associated with each of the multiple location identifiers to further determine relevancy to the semantic type;

ranking each of the multiple location identifiers based on the relevancy determined from the data mining.

27. The method of claim 26, further comprising, re-defining the location identifier pattern based on the relevancy to the semantic type determined from the data mining of the multiple location identifiers.

28. The method of claim 12, wherein, the location identifier pattern is manually specified or automatically extracted using predetermined rules.

29. The method of claim 12, further comprising,

analyzing additional sitemaps of other domains;

identifying, from the additional sitemaps, multiple additional location identifiers of the other domains matching the location identifier pattern defined for the semantic type;

indexing each of the additional multiple location identifiers as being associated with objects having content of the semantic type.

30. The method of claim 12, wherein, the location identifier pattern includes a wildcard segment.

31. The method of claim 30, wherein, the location identifier pattern further includes a domain name segment.

32. The method of claim 12, further comprising,

defining a supplemental location identifier pattern for identifying objects having content related to an attribute of the semantic type.

33. The method of claim 32, further comprising,

identifying, from the sitemap, multiple supplemental location identifiers of the particular domain matching the supplemental location identifier pattern defined for the attribute of the semantic type;

indexing each of the multiple supplemental location identifiers as being associated with objects having content related to the attribute of the semantic type.

34. A system for generating user-customized search results using user-defined semantic types, the system, comprising:

a user module that tracks a first set of semantic types defined by a first user and a second set of semantic types defined by a second user;

a recognition module that identifies a first set of URI patterns that are associated with the first set of semantic types defined by the first user and a second set of URI patterns that are associated with the second set of semantic types defined by the second user;

a search optimizer module that semantically categorizes a first set of search results for the first user, as having content related to one or more of the first set of semantic types defined by the first user;

the search optimizer module further semantically categorizes a second set of search results for the second user, as having content related to one or more of the second set of semantic types defined by the second user;

wherein, the first set of search results are categorized using the first set of URI patterns;

wherein, the second set of search results are categorized using the second set of URI patterns.

35. The system of claim 34, further comprising:

a dialogue analyzer module that parses a web-dialogue regarding a particular webpage in which the first user is participating according to a grammar;

the search optimizer module further semantically categorizes the particular web page, based on the parsing, as having content related to one or more of the first set of semantic types defined by the first user.

36. The system of claim 34,

wherein, the first set of search results are identified in response to a keyword search requested by the first user;

wherein, the second set of search results are identified in response to the keyword search requested by the second user that is a same keyword search as the first user'"'"'s keyword search.

37. The system of claim 34, wherein:

the pattern recognition module identifies a first additional set of URI patterns that are associated with a first set of attributes of a semantic type that are defined by the first user;

the search optimizer module semantically categorizes the first set of search results for the first user, as having content related to one or more of the first set of attributes of the semantic type;

wherein, the first set of search results are categorized using the first additional set of URI patterns;

wherein, the first additional set of URI patterns are identified from a first additional set of URIs of web pages having content specified by the first user as having content related to one or more of the first set of attributes of the semantic type.

38. A machine-readable storage medium having stored thereon a set of instructions which when executed perform a method for building a semantics-enhanced search engine, the method, comprising:

defining a URI pattern for identifying web pages having content of a semantic type;

wherein, the URI pattern includes a wildcard segment;

analyzing a sitemap of a particular domain;

identifying, from the sitemap, multiple URIs of the particular domain matching the URI pattern defined for the semantic type;

tagging each of the multiple URIs as web pages having content of the semantic type;

defining a supplemental URI pattern for identifying web pages having content related to an attribute of the semantic type;

identifying, from the sitemap, multiple supplemental URIs of the particular domain matching the supplemental URI pattern defined for the attribute of the semantic type;

tagging each of the multiple supplemental URIs as web pages having content related to the attribute of the semantic type.