METHOD AND SYSTEM FOR BUILDING TEXT DESCRIPTIONS IN A SEARCH DATABASE

• US 20090132514A1
• Filed: 11/16/2007
• Published: 05/21/2009
• Est. Priority Date: 11/16/2007
• Status: Abandoned Application

• Current Assignee: IAC Search & Media Incorporated (Match Group, Inc.)
• Original Assignee: IAC Search & Media Incorporated (Match Group, Inc.)
• Inventors: Vegnaduzzo, Stefano, Kislyuk, Oleg S., Yang, Yue Rona
• Application Number:
  US11/941,877
• Publication Number:
  US 20090132514A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Abstract

The invention provides for a method of building a search database including associating a plurality of text descriptions associated with a first set of locations with a first category associated with all of the locations of the first set to build a first combined text description, and associating the combined text description with each one of the locations of the first set.

Claims

1. A method of building a search database, comprising:

associating a plurality of text descriptions associated with a first set of locations with a first category associated with all of the locations of the first set to build a first combined text description; and

associating the combined text description with each one of the locations of the first set.

2. The method of claim 1, further comprising:

associating a plurality of text descriptions associated with a second set of locations with a second category associated with all of the locations of the second set to build a second combined text description; and

associating the combined text description with each one of the locations of the second set, wherein at least some of the locations of the second set do not belong to the first set of locations.

3. The method of claim 2, wherein at least some of the locations of the second set belong to the first set of locations.

4. A system for information retrieval, comprising:

a reception component that receives at least one data entry from at least one data source, including web-crawled data;

a processing component to determine the classifications of the at least one data en try;

a building component that builds at least one component of the language model associated to the at least one data entry; and

a merging component that builds a final language model associated to the at least one data entry using the at least one component of the language model.

5. The system of claim 4, further comprising a ranking component that uses the final language model to estimate the relevance of the at least one data entry in a user query.

6. The system of claim 4, wherein the building component builds the at least one component of the language model using text information from data possessing at least one of the same classifications as the at least one data entry.

7. The system of claim 4, wherein the merging component performs a linear combination of the at least one component of the language model to build the final language model.

8. A computer-readable medium, having stored thereon a set of instructions which, when executed by a processor of a computer, executes a method of building a search database, comprising:

associating a plurality of text descriptions associated with a first set of locations with a first category associated with all of the locations of the first set to build a first combined text description; and

associating the combined text description with each one of the locations of the first set.