Method for electronically recognizing and parsing information contained in a financial statement

Abstract

A method for processing a file containing a financial statement in uncoded format such as a financial statement stored in an ASCII file. The starting locations of the various types of financial tables in the financial statement as indicated by their table titles are identified by grammar descriptors. When all the table titles in the file are identified, a table title is then selected for processing. Typically after the table title, there are the associated column headers for the table, and they are analyzed and determined. Following the column headers, there are lines of text that are separated into line items, where each line item is composed of a label and/or one or more numbers corresponding to the label. With these line items, the next task is to parse these line items to verify that these line items make up the table and to identify the components of the financial table. A financial table is comprised of line items, subtotals, and grand totals. By recognizing the mathematical structure of the table, the relationship between the line items, subtotals, and grand totals can be readily understood and categorized.