Intelligent query system and method using phrase-code frequency-inverse phrase-code document frequency module

• US 20060031218A1
• Filed: 04/22/2005
• Published: 02/09/2006
• Est. Priority Date: 07/22/2004
• Status: Active Grant

• Current Assignee: Factiva Incorporated
• Original Assignee: Factiva Incorporated
• Inventors: Cipollone, Peter M.
• Granted Patent: US 7,698,333 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/367 Ontology

Abstract

An intelligent query system and method used in a search and retrieval system provides an end-user the most relevant, meaningful, up-to-date, and precise search results. The system and method allows an end-user to benefit from an experienced recommendation that is tailored to a specific industry. For example, the system and method recognizes that the phrases “strike outs” and “home run” are much more strongly correlated with “BASE” as opposed to “EQUITIES.” When a search is conducted or a lookup is done in a map, the system and method recommends the strongest correlation as “BASE.”

Claims

1. An intelligent query method, comprising the steps of:

providing a plurality of documents which contain multimedia contents;

categorizing each of the documents into a taxonomy with corresponding taxonomy elements wherein the taxonomy is pre-defined;

filtering/transforming the multimedia contents and discarding a portion of the taxonomy elements;

storing the filtered/transformed multimedia contents in a database; and

calculating a correlation value of the filtered/transformed multimedia contents.

2. An intelligent query method used in a search and retrieval system, comprising the steps of:

providing a plurality of documents which contain multimedia contents including text;

categorizing each of the documents into a taxonomy with corresponding taxonomy elements wherein the taxonomy is pre-defined;

filtering terms within the text to generate terms (Tt) and stop terms (Ts), wherein terms (Tt) are single words which express semantic value to the document, and stop terms (Ts) are single words which express no semantic value;

discarding the stop terms (Ts) and defining the remaining terms (Tt) as T;

transforming the terms (T) to eliminate multi-collinearity and correlating the transformed terms t to each taxonomy element c on a containing document, wherein t is an element of T, and c is an element of C;

storing t and c in a database;

counting the documents that contain c; and

increasing a correlation value between term t and taxonomy element c each time when the term t appears in the document.

3. The method of claim 2, further comprising a step of generating an IQ map, which comprises:

scoring t-c pairs according to a PCF-IPCDF scoring system or model;

loading the t-c pairs with the highest scores into a map structure for facilitating lookup of the taxonomy element c from the term element t; and

deducing the taxonomy element c from the term element t.