This talk demonstrates workflows to incorporate text data into actuarial classification and regression tasks. The main focus is on methods employing transformer-based models. The talk provides practical approaches to handle classification tasks in situations with no or only few labeled data. A dataset with short property insurance claims descriptions is used to demonstrate the techniques. The results achieved by using the language-understanding skills of off-the-shelf natural language processing (NLP) models with only minimal pre-processing and fine-tuning clearly demonstrate the power of transfer learning for practical applications.  

This case study has been done as part of the “Data Science” working group of the Swiss Association of Actuaries (SAA). The group publishes tutorials that discuss the use of machine learning techniques for actuarial applications. The tutorials are self-explanatory and its code and data is publicly available on the website www.actuarialdatascience.org.