Intelligent search and retrieval system and method

• US 7,836,083 B2
• Filed: 02/18/2005
• Issued: 11/16/2010
• Est. Priority Date: 02/20/2004
• Status: Active Grant

• Current Assignee: Factiva Incorporated, Dow Jones Reuters Business Interactive LLC (News Corporation)
• Original Assignee: Factiva Incorporated
• Inventors: Cipollone, Peter M.
• Primary Examiner(s): Mofiz; Apu M
• Assistant Examiner(s): Rahman; Mohammad N
• Application Number:
  US11/060,928
• Publication Number:
  US 20050187923A1
• Time in Patent Office: 2,097 Days
• Field of Search: 707 2- 5, 707/999.2, 707/792
• US Class Current: 707/792
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence

Abstract

An intelligent search and retrieval system and method is provided to allow an end-user effortless access yet most relevant, meaningful, up-to-date, and precise search results as quickly and efficiently as possible. The method may include providing a query profiler having a taxonomy database; receiving a query from a user; accessing the taxonomy database of the query profiler to identify a plurality of codes that are relevant to the query; augmenting the query using the codes to generate feedback information to the user for query refinement, the feedback information including a plurality of query terms associated with the query and to be selected by the user; presenting the feedback information to the user; receiving one of the query terms from the user; and identifying a source of the query term and presenting to the user. The system may include a query profiler having a taxonomy database to be accessed upon receiving a query from a user, which identifies a plurality of codes that are relevant to the query; means for augmenting the query using the codes to generate feedback information to the user for query refinement, the feedback information including a plurality of query terms associated with the query and to be selected by the user; and means for identifying a source of the query term, upon receiving one of the query terms from the user.

Claims

1. An intelligent search and retrieval method, comprising the steps of:

providing a query profiler having a taxonomy database, the taxonomy database including a plurality of taxonomy codes which have explicitly defined contextual relationship and are semantically related;

receiving a query from a user;

accessing the taxonomy database of the query profiler to identify the taxonomy codes that are relevant to the query;

wherein taxonomy codes are identified using a phrase-code frequency-inverse phrase-code document frequency (pcf-ipcdf) score;

wherein phrase-code frequency, pcf(p,c), is defined as a number of times a phrase p appears in one or more categorized documents containing a code c;

wherein inverse phrase-code document frequency, ipcdf, is defined as the logarithm of;

a number of the documents coded with code c, D(c), divided by a number of the documents for which the phrase p and code c appear together, df(p,c);

wherein the pcf-ipcdf score, s(p,c), is defined as pcf(p,c) multiplied by ipcdf(p,c);

augmenting the query using the taxonomy codes;

generating feedback information to the user for query refinement, the feedback information including a plurality of query terms associated with the query and to be selected by the user;

presenting the feedback information to the user;

receiving one of the query terms from the user; and

identifying a source of the query term and presenting to the user;

wherein the taxonomy database is generated by;

parsing the natural language from the one or more categorized documents;

parsing one or more associated taxonomy codes into a data structure;

filtering unnecessary or undesirable code elements from the data structure;

extracting phrases from the text of the one or more categorized documents;

sorting and collating the extracted phrases into a counted phrase list; and

mapping the counted phrase list and the one or more associated taxonomy codes into a data table.

2. The method of claim 1, wherein the taxonomy database of the query profiler comprises a timing identifier for identifying a timing rage, wherein the method further comprises receiving the query with a time range and identifying the source of the query term with the time range.

3. The method of claim 1, wherein the taxonomy database of the query profiler comprises a query term ranking module, wherein the module provides a relevance score corresponding to the number of times the query term appears in the documents containing the corresponding taxonomy code and the number of documents for which the query term and the corresponding taxonomy code appear together.