Machine learning for the analysis of qualitative financial reports

The Swedish Financial Supervisory Authority (SFSA) conducts risk-based supervision of the Swedish insurance market through a wide range of activities. One of the activities is a qualitative assessment of the financial reports submitted by insurance companies. The set of reports is extensive both in length and type, which makes the assessment process time-consuming. The overarching goal of this project is to investigate the use of Machine Learning methods for identifying information in the set of qualitative financial reports, that can serve as useful insight for the SFSA officers and help them prioritise their reading order.