System and methods for improved linguistic pattern matching

• US 20020099536A1
• Filed: 09/21/2001
• Published: 07/25/2002
• Est. Priority Date: 09/21/2000
• Status: Active Grant

• Current Assignee: JP Morgan Chase & Co, Vastera, Inc. (JP Morgan Chase & Co), Livingston International Professional Services, LLC
• Original Assignee: Vastera, Inc. (JP Morgan Chase & Co)
• Inventors: Finney, Michael, Cobey, Derrick, Bordner, Ty
• Granted Patent: US 7,328,211 B2
• Time in Patent Office: Days
• US Class Current: 704/10
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

A system and method for reducing the number of false negatives with minimal impact on false positive search results while allowing searches to return phonetic equivalents, misspellings, common short names, and other such applicable information. Also disclosed are new, N-Gram based indexing and search systems and methods which facilitate searching of data containing non-Arabic letters, such as numbers, symbols, and foreign language characters. The present invention further allows indexing and other features to be customized, thereby further enhancing search results. Further, the present invention provides a system for improving linguistic pattern matching search results based on dynamically modified search attributes.

Claims

1. A method for comparing a query against data contained within a database comprising the steps of:

(a) receiving said query;

(b) extracting a plurality of attributes from said query;

(c) converting said plurality of attributes from said query, using at least one linguistic pattern matching analytical tool, into a plurality of linguistic pattern strings;

(c) comparing said plurality of linguistic pattern strings with at least one stored linguistic pattern string from at least one stored attribute contained within said database for providing a set of matches;

(d) analyzing each match of said set of matches, using said at least one linguistic pattern matching analytical tool, to provide at least one set of matched attributes; and

(e) combining all of said at least one set of matched attributes to provide a combined result.

2. The method of claim 1, further including the step of filtering said combined result according to pre-selected criteria.

3. The method of claim 2, further including the step of filtering said combined result according to matching a selected attribute.

4. The method of claim 1, further comprising the step of employing a Metaphone based analysis, a Phonex based analysis, a Soundex based analysis, an N-gram based analysis, an edit-distance based analysis and a dictionaries based analysis.

5. A method of comparing query information about a party against a plurality of restricted parties information contained within a database comprising the steps of:

(a) receiving said query about said party;

(b) extracting a plurality of attributes from said query;

(c) converting said plurality of attributes from said query, using at least one linguistic pattern matching analytical tool, into a plurality of linguistic pattern strings;

(d) comparing said plurality of linguistic pattern strings with at least one stored linguistic pattern string from said plurality of restricted parties information contained within said database for providing a set of matches;

(d) analyzing each match of said set of matches, using said at least one linguistic pattern matching analytical tool, to provide at least one set of matched attributes; and

(e) combining all of said at least one set of matched attributes to provide a combined result.

6. The method of claim 5, further including the step of filtering said combined result according to pre-selected criteria.

7. The method of claim 6, further including the step of filtering said combined result according to matching a selected attribute.

8. The method of claim 5, further comprising the step of employing a Metaphone based analysis, a Phonex based analysis, a Soundex based analysis, an N-gram based analysis, an edit-distance based analysis and a dictionaries based analysis.

9. A system for comparing a query against data contained within at least one database comprising:

(a) a central processing unit having at least one electronic communications port for receiving said query, wherein said central processing unit is attached to said at least one database;

(b) at least one extraction tool accessible to said central processing unit for extracting a plurality of attributes from said query;

(c) at least one linguistic pattern analytical tool accessible to said central processing unit for converting said plurality of attributes from said query into a plurality of linguistic pattern strings, and for comparing said plurality of linguistic pattern strings with at least one stored linguistic pattern string contained within at least one of said database for providing a set of matches;

(d) said at least one linguistic pattern analytical tool accessible to said central processing unit for analyzing each match of said set of matches to provide at least one set of matched attributes; and

(e) at least one combining tool accessible to said central processing unit for combining all of said at least one set of matched attributes to provide a combined result.

10. The system of claim 9, further comprising at least one filtering tool accessible to said central processing unit for filtering said combined result according to pre-selected criteria.

11. The system of claim 10, wherein said at least one filtering tool filters said combined result according to matching a selected attribute.

14. The computer program product of claim 13, further comprising computer readable code for filtering said combined result according to pre-selected criteria.

15. The computer program product of claim 14, wherein said computer readable code for filtering filters said combined result according to matching a selected attribute.

16. The computer program product of claim 13, wherein said computer readable program code for converting a plurality of information from said query into a plurality of linguistic pattern strings and for comparing said plurality of linguistic pattern strings with at least one stored linguistic pattern string contained within said database to provide a set of matches is comprised of Metaphone based analysis code, Phonex based analysis code, Soundex based analysis code, N-gram based analysis code, edit-distance based analysis code and dictionaries based analysis code.

12. The system of claim 105, wherein said at least one linguistic pattern analytical tool is comprised of a Metaphone based analysis, a Phonex based analysis, a Soundex based analysis, an N-gram based analysis, an edit-distance based analysis and a dictionaries based analysis.

13. A computer program product for querying a database comprising:

a computer useable medium having a computer readable program code embodied in the computer usable medium for causing an application program to execute on a computer system, the computer readable program code comprising;

computer readable program code for receiving a query;

computer readable program code for extracting a plurality of attributes from said query;

computer readable program code for at least one linguistic pattern analytical tool for converting a plurality of information from said query into a plurality of linguistic pattern strings;

computer readable program code for comparing said plurality of linguistic pattern strings with at least one stored linguistic pattern string contained within said database to provide a set of matches;

computer readable program code for analyzing each match of said set of matches to provide at least one set of matched attributes; and

computer readable program code for combining all of said at least one set of matched attributes to provide a combined result.

17. A method for comparing a query against data contained within a database comprising the steps of:

(a) receiving said query;

(b) extracting a plurality of attributes from said query;

(c) converting said plurality of attributes, using a Metaphone based linguistic pattern analytical tool, into a plurality of Metaphone linguistic pattern strings;

(d) comparing at least one of said plurality of Metaphone linguistic pattern strings with said at least one stored linguistic pattern string contained within said database to provide plurality of Metaphone matches;

(e) converting said plurality of attributes, using a Phonex based linguistic pattern analytical tool, into a plurality of Phonex linguistic pattern strings;

(f) comparing at least one of said plurality of Phonex linguistic pattern strings with said at least one stored linguistic pattern string contained within said database to provide a plurality of Phonex matches;

(g) converting said plurality of attributes, using a Soundex based linguistic pattern analytical tool, into a plurality of Soundex linguistic pattern strings;

(h) comparing at least one of said plurality of Soundex linguistic pattern strings with said at least one stored linguistic pattern string contained within said database to provide a plurality of Soundex matches;

(i) converting said plurality of attributes, using an N-gram based linguistic pattern analytical tool, into a plurality of N-gram linguistic pattern strings;

(j) comparing at least one of said plurality of N-gram linguistic pattern strings with at least one stored linguistic pattern string contained within said database to provide a plurality of N-gram matches;

(k) combining said plurality of Metaphone matches, said plurality of Phonex Matches, said plurality of Soundex matches, and said plurality of N-gram matches to form a set of matches;

(l) analyzing each match of said set of matches using said Metaphone based linguistic pattern analytical tool, Phonex based linguistic pattern analytical tool, said Soundex based linguistic pattern analytical tool, an edit-distance based linguistic pattern analytical tool, and a dictionaries based linguistic pattern analytical tool to provide at least one set of matched attributes; and

(m) combining all of said at least one set of matched attributes to provide a combined result.