A key part of this approach is matching the form data to the list of publicly traded companies. Unfortunately, there can be different entries for any particular stock (e.g. “General Electric”, “General Electric Co.”, and “General Electric Company”). Fortunately, there is a great Python package called fuzzywuzzy (great name) that performs well at fuzzy string matching. In this way, you can extract company names from the forms with the added benefit of stock name standardization.