Process for determination of text relevancy

• US 5,694,592 A
• Filed: 08/28/1995
• Issued: 12/02/1997
• Est. Priority Date: 11/05/1993
• 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, Jim
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Lewis, C.
• Application Number:
  US08/520,027
• Time in Patent Office: 827 Days
• Field of Search: 395/600, 395/604, 395/603, 395/605, 395/759, 395/793, 364/300
• US Class Current: 1/1
• 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: G06F 16/353 into predefined classes
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

This is a procedure for determining text relevancy and can be used to enhance the retrieval of text documents by search queries. This system helps a user intelligently and rapidly locate information found in large textual databases. A first embodiment determines the common meanings between each word in the query and each word in the document. Then an adjustment is made for words in the query that are not in the documents. Further, weights are calculated for both the semantic components in the query and the semantic components in the documents. These weights are multiplied together, and their products are subsequently added to one another to determine a real value number(similarity coefficient) for each document. Finally, the documents are sorted in sequential order according to their real value number from largest to smallest value. Another, embodiment is for routing documents to topics/headings (sometimes referred to as faltering). Here, the importance of each word in both topics and documents are calculated. Then, the real value number (similarity coefficient) for each document is determined. Then each document is routed one at a time according to their respective real value numbers to one or more topics. Finally, once the documents are located with their topics, the documents can be sorted. This system can be used to search and route all kinds of document collections, such as collections of legal documents, medical documents, news stories, and patents.

Claims

1. A computer implemented method of creating similarity coefficients between sequences of words in documents that are being searched in a database by a natural word query without parsing the query words nor the words in the documents, and without removing any of the query words and any of the words in the documents, the method comprising the steps of:

(a) branching out the meanings of each and every word in a natural word query into respective probabilities of occurrence for each of the meanings in the natural word query;

(b) branching out the meanings of words in a document searched by the natural word query into respective probabilities of occurrence for each of the meanings of the words in each of the documents;

(c) determining a similarity coefficient between the probabilities of occurrence of words in the natural language query and the probabilities of occurrence of the words in the document;

(d) repeating steps (a) to (c) for each additional document searched by the natural language query; and

(e) ranking the documents being searched in order of their similarity coefficients without parsing of the natural language query and the documents, and without removing any words from the natural language query nor from the documents.

2. The computer implemented method of creating similarity coefficients between sequences of words of claim 1, wherein the meanings of a word further include definitions of words.

3. A computer implemented method of creating similarity coefficients between sequences of words of claim 1, wherein the meanings of a word further include senses.

4. A computer implemented method of creating similarity coefficients between sequences of words of claim 1, wherein the meanings of a word further include categories.

5. The computer implemented method of creating similarity coefficients between sequences of words of claim 4, wherein the categories further includes:

a semantic lexicon of categories.

6. The computer implemented method of creating similarity coefficients between sequences of words of claim 5, wherein step(a) further includes:

determining a probability value for each word in the query matching the semantic categories; and

wherein step(b) further includes;

determining a probability value for each word in the document matching the semantic categories.

7. The computer implemented method of creating similarity coefficients between sequences of word of claim 6, wherein step(c) of determining similarity coefficients further includes:

(i) calculating weights of a semantic component in the query based on the probability values of the words in the query;

(ii) calculating weights of a semantic component in the document based on the probability values of the words in the document;

(iii) multiplying query component weights by document component weights into products; and

(iv) adding the products together to represent the similarity coefficient as a real-value number for the document.

8. The computer implemented method of creating similarity coefficients between sequences of word of claim 1, wherein each document is chosen from at least one of:

a word, a sentence, a line, a phrase and a paragraph.

9. A computer implemented method of creating similarity coefficients between sequences of words in documents that are being searched in a database by a natural word query without parsing the query words nor the words in the documents, and without removing any of the query words and any of the words in the documents, the method comprising the steps of:

(a) branching out the meanings of each and every word in a natural word query into respective probabilities of occurrence for each of the meanings in the natural word query, wherein the query includes at least one word;

(b) branching out the meanings of each and every word in a document searched by the natural word query into respective probabilities of occurrence for each of the meanings of the words in each of the documents, wherein the document includes at least one word;

(c) determining a similarity coefficient between the probabilities of occurrence of words in the natural language query and the probabilities of occurrence of the words in the document;

(d) repeating steps (a) to (c) for each additional document searched by the natural language query; and

(e) ranking all the documents being searched in order of their similarity coefficients without parsing of the natural language query and the documents, and without removing any words from the natural language query nor from the documents.