On-line iterative multistage search engine with text categorization and supervised learning

• US 20080168056A1
• Filed: 01/09/2008
• Published: 07/10/2008
• Est. Priority Date: 01/09/2007
• Status: Active Grant

• Current Assignee: Favoweb, Ltd.
• Original Assignee: Favoweb, Ltd.
• Inventors: Porotsky, Sergey, Bluvband, Zigmund
• Granted Patent: US 8,150,822 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages

Abstract

A system and method for performing a user supervised search which uses algorithm with parameters that are updated based on the review of the search results by the user.

Claims

1. A method comprising:

(a) receiving an initial list of search results in response to at least one search term;

(b) ranking by a user the appropriateness of at least one of said results;

(c) updating text mining algorithm parameters according to said ranking (d) performing next search using said updated text mining algorithm parameters;

(e) providing an updated list of search results; and

(f) repeating steps (b) to (e) until said user stops said next search.

2. The method of claim 1 wherein said at least one search term is included in a search term group, and

wherein the size of said search term group is reduced by excluding repeating terms which include at least one of the following list:

suffix, prefix.

3. The method of claim 1 wherein said step of updating text mining algorithm parameters comprise setting a threshold to distinguish between suitable and non-suitable documents.

4. The method of claim 1, wherein said step of receiving an initial list of search results is initiated automatically, based on a pre-prepared set of search terms.

5. The method of claim 1, wherein said ranking is a binary ranking.

6. The method of claim 1, wherein said ranking is a multi-level ranking.