EdgarScan

EDGARSCAN is a program designed to read and analyze SEC EDGAR filings. While there are many kinds of documents that public companies file with the SEC, we are mostly concerned with the quarterly and annual filings (10-Qs and 10-Ks) which contain key financial tables such as Income Statements and Balance Sheets. The filings are "Semi-Structured" documents. Documents are structured along loose guidelines set forth by the Securities and Exchange Commission (e.g. what sections must be present in the filing, and what accounting information must be present), but in practice there is such variety in the presentation of the financial tables that it is quite difficult to extract useful information automatically.