Method and system for using query information to enhance catergorization and navigation within the whole knowledge base

• US 7,464,074 B2
• Filed: 11/10/2003
• Issued: 12/09/2008
• Est. Priority Date: 11/13/2002
• Status: Expired due to Fees

• Current Assignee: Moshe Mizrahi
• Original Assignee: Insight Look Ltd.
• Inventors: Sebbane, Danny
• Primary Examiner(s): Rones; Charles
• Assistant Examiner(s): Khoshnoodi; Fariborz
• Application Number:
  US10/535,665
• Publication Number:
  US 20060053142A1
• Time in Patent Office: 1,856 Days
• Field of Search: 707/3, 707/2, 707/200, 707/100, 705/14
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2423 Interactive query statement specification based on a database schema
• CPC Class Codes: G06F 16/2428 Query predicate definition using graphical user interfaces, including menus and forms G06F16/2423 takes precedence
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/903 Querying for retrieval from the web G06F16/953
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A method is disclosed to create some structure from the knowledge base of an organization, the knowledge base including a document database (DB) and queries submitted by users concerning the documents, wherein the method performs monitoring and clustering to enable analysis of the ideas of the organization. The first step of the method involves users submitting queries by entries on a graphical user interface (GUTI). Subsequent steps include recording the queries of the organization, which includes saving submitter and organization information, comparing queries using a weight matrix generated by a distance function and clustering of the queries into a semantic structure.

Claims

1. A computer implemented method to create a structure from a knowledge base of an organization, the knowledge base being part of a system comprising a document database (DB) and queries submitted by users concerning the documents, wherein the method performs definitive and decisive matching and grouping the queries into clusters to enable analysis of the knowledge base of the organization;

the method comprising;

submitting by the users in the organization of queries having weighted characteristics;

recording the details of the submitter comprising at least;

job title;

department name; and

employee contact references comprising at least email address and level of security authorization;

comparing queries using a weight matrix generated by a distance function;

clustering of the queries into a semantic structure based on said weight matrix, by grouping said queries into a ‘

prioritized structure’

based on the comparison of a location of a word in said query to the location of the same word in another query; and

rating of a new query relative to the nearest of said clusters, wherein said new queries are evaluated in one of real-time and periodically, to determine whether to one of;

add said query to an existing cluster; and

form a separate “

satellite”

cluster.

2. The method according to claim 1, further comprising:

periodically updating the gathering of data of said newly submitted queries and resulting documents into a sum total of data elements for the organization;

entering said sum total in the organization DB without prior categorization of said sum total by subject matter;

instantly performing a matching procedure of assembling queries and documents keywords into clusters;

categorizing said newly submitted queries according to a continuously updated list of categories; and

repeatedly redefining categories and clusters according to new queries and documents.

3. The method according to claim 2, comprising the steps of:

gathering data into the organization DB;

generating a vector structure of the data; and

using the vector structure in order to form semantic familiarities (clustering words, i.e., “

connections”

).

4. The method according to claim 2, further comprising enhancing the queries for later pre-processing of the data, in order to best exploit the latter element of the method.

5. The method according to claim 4, wherein enhancing comprises:

enhancing words appearing in queries by multiplying the number of appearances with a constant;

comparing the distribution of a word within the organization DB and distribution of said word in Common Language (CL); and

weighting words appearances in the DB and the queries relative to appearances in the (CL).

6. The method according to claim 4, further comprising clustering the data.

7. The method according to claim 6, wherein clustering the data comprises:

using information theories in order to assemble and represent the data;

using queries as prior knowledge for the algorithms processing the data;

clustering data (agglomerative, sequential clustering); and

using queries as a predisposed factor, thereby replacing the random factor when performing clustering.

8. The method according to claim 6, further comprising using queries'"'"' data for searching information (implementing a search engine).

9. The method according to claim 8, comprising the steps of:

searching information using the queries'"'"' structure (clusters);

presenting queries'"'"' structure with respect to a new query (when a user presents a new query, the system rates the nearest clusters according to the new query); and

presenting submitted queries in order to facilitate the submission of a new query.

10. The method according to claim 8, further comprising using the queries structure to create an organization map.

11. The method according to claim 1, wherein using the queries structure to create an organization map comprises:

developing amethod that facilitates the designation of experts concerning the requested data; and

providing a graphical organization map of the data occurrences and the experts.