Robert P Gay

US Patent:

6792145, Sep 14, 2004

Filed:

Jun 8, 2001

Appl. No.:

09/877836

Inventors:

Robert W. Gay - New York NY 10023

International Classification:

G06K 446

US Classification:

382190, 382209, 705 30, 715503

Abstract:

The present invention is directed to a method for extracting textual as well as tabular data material from financial documents. Initially, a comparison is made to determine the type of data schedule material provided in the document. Subsequently, the character strings of the financial document are compared to character strings provided in previous documents or in various databases. The database of the previous document would include the textual material in a first plane, and the tabular material also in that first plane. If a character string match is made between a new document and an old document, the new tabular data material would be provided in a data matrix in a second plane but the corresponding textual material would not be included in the textual matrix provided in that second plane. Only character strings not matched in the textual material of the first plane would be provided in the textual matrix of the second plane.