Natural language method and system for searching for and ranking relevant documents from a computer database

• US 6,088,692 A
• Filed: 04/05/1999
• Issued: 07/11/2000
• Est. Priority Date: 12/06/1994
• Status: Expired due to Term

• Current Assignee: University of Central Florida Research Foundation Inc. (State University System of Florida)
• Original Assignee: University of Central Florida (State University System of Florida)
• Inventors: Driscoll, James R.
• Primary Examiner(s): HOMERE, JEAN RAYMOND
• Application Number:
  US09/281,917
• Time in Patent Office: 463 Days
• Field of Search: 707/3, 707/5, 707/500, 707/9, 704/9
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3334 Selection or weighting of terms from queries, including natural language queries
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99939 Privileged access

Abstract

Natural Language search system and method for retrieving relevant documents from a text data base collection comprised of patents, medical and legal documents, journals, news stories and the like. Each small piece of text within the documents such as a sentence, phrase and semantic unit in the data base is treated as a document. Natural language queries are used to search for relevant documents from the data base. A first search query creates a selected group of documents. Each word in both the search query and in the documents are given weighted values. Combining the weighted values creates similarity values for each document which are then ranked according to their relevant importance to the search query. A user reading and passing through this ranked list checks off which document are relevant or not. Then the system automatically causes the original search query to be updated into a second search query which can include the same words, less words or different words than the first search query. Words in the second search query can have the same or different weights compared to the first search query. The system automatically searches the text data base and creates a second group of documents, which as a minimum does not include at least one of the documents found in the first group. The second group can also be comprised of additional documents not found in the first group. The ranking of documents in the second group is different than the first ranking such that the more relevant documents are found closer to the top of the list.

Claims

1. A natural language search method for retrieving relevant information from a computer database collection, comprising the steps of:

(a) searching a database collection in a computer using a first search query of natural language to retrieve a first group of selected small pieces of information each being dissimilar from one another;

(b) ranking each of the selected small pieces of information into a first relevant ranked list of the selected small pieces of information;

(c) applying feedback information based on a subsequent determination of the relevancy of each of the small pieces of information in the first ranked list to create a second query different from the first query; and

(d) repeating steps (a) to (b) to form a second relevant ranked list of the selected small pieces of information.

2. The natural language search method of claim 1, further including the step of:

retrieving the first relevant ranked list and the second relevant ranked list without annotating, parsing or pruning.

3. The natural language search method of claim 1, further comprising the steps of:

(e) repeating step (c) to create a third query different from the first query and the second query; and

(f) repeating step (d) to form a third relevant ranked list of the selected small pieces of information.

4. The natural language search method of claim 3, further comprising the steps of:

(g) repeating step (c) to create a fourth query different from the first query, the second query and the third query;

(h) repeating step (e) to form a fourth relevant ranked list of the selected small pieces of information.

5. A natural language search method for retrieving relevant information from a computer database, comprising the steps of:

(a) searching a computer database using a first natural language search query to retrieve a first group of selected portions of information each of the portions of information being dissimilar from one another;

(b) ranking each of the selected portions of information in a first relevant ranked list of the selected portions of information;

(c) applying feedback information based on a subsequent determination of the relevancy of each of the portions of information in the first ranked list to create a second search query different from the first search query; and

(d) searching the computer database using the second search query to retrieve a second group of the selected portions of information and ranking the second group in a second relevant ranked list of the selected portions of information.

6. A natural language search method for retrieving relevant information from a computer data base, comprising the steps of:

searching a computer database using a first natural language search query to retrieve a first group of selected portions of information, each of the portions of information being different from one another;

ranking each of the selected portions of information in a first relevant ranked list of the selected portions of information;

applying feedback information based on a subsequent determination of the relevancy of each of the portions of information in the first ranked list to create a second search query different from the first search query; and

searching the computer database using the second search query to retrieve a second group of the selected portions of information.