Impact of Text Mining Application on Financial Footnotes Analysis Research in Progress

In recent decade and with the advent of the eXtensible Business Reporting Language (XBRL), financial reports have a great mutation in terms of a unified reporting process. Nevertheless, the unstructured part of financial reports, so called footnotes, remains as barrier facing an accurate automatic and real-time financial analysis. The purpose of this paper is to investigate whether the text mining approach is an appropriate solution to assist analyzing textual financial footnotes or not. The implemented text mining prototype is able to classify textual financial footnotes into related pre-defined categories automatically. This avoids manually reading of the entire text. Different text classification supervised algorithms have been compared, where the decision tree by 90.65% accuracy performs better rather than other deployed classifiers. This research provides preliminary insights about the impact of using a text mining approach on automatic financial footnote analysis in terms of saving time and increasing accuracy.