Statistical Learning in Actuarial Applications Working Party
Learn more about our latest working party formed to explore statistical learning in actuarial applications with the aim to construct highly flexible actuarial models using copulas, dependence modelling and regression-based models.
Featured articles of the month
Python Visualisation Dashboard
Try out the latest python visualisation dashboard created by Afzaal Ahmed that explores claims from the French motor third-party liability (MTPL) insurance portfolio.
All together now: modelling claims using federated learning
Read more as Małgorzata, Dylan and Claudio discuss federated learning and how to model claims anonymously using this new technique.
Originally published by The Actuary, September 2021. © The Institute and Faculty of Actuaries.
Supervised learning techniques in claims frequency modelling
Read on as Neptune Jin shares the Data Science Research Section’s work looking at the merits of different supervised learning techniques for claim frequency modelling.
Originally published by The Actuary, July 2021. © The Institute and Faculty of Actuaries.
Insurance: Collaboration without compromise
Learn more as Małgorzata Śmietanka introduces the opportunities for federated learning and privacy-preserving data access in insurance.
Originally published by The Actuary, March 2021. © The Institute and Faculty of Actuaries.
A little bird told me...
John Ng and Melanie Zhang discuss how they analysed Twitter sentiment relating to COVID-19 and insurance – and how insurers could use such analysis to uncover insights pertinent to public interest and the insurance industry.
Originally published by The Actuary, February 2021. © The Institute and Faculty of Actuaries.
Featured working parties
In this working party, we investigate tools and packages available to perform data visualisation and produce outputs that conform to best practice in the area. Click on the link below to find out more about our work.
Natural Language Processing
Find out more on how we classified sentiment of UK Twitter messages on COVID-19 using an end-to-end Natural Language Processing (NLP) pipeline and supervised Machine Learning techniques.
Federated learning is a pioneering privacy-preserving data technology and also a new machine learning model trained on distributed data sets, removing the need to store sensitive training data on a central server.