Data Science Working Party - Research

The Research Section consists of an international group of actuaries, data scientists, statisticians and academics bringing together multiple domain expertise to research and develop data science techniques in actuarial applications.

We are open to new volunteers with a general understanding of data science and have time to contribute. Non-members of IFoA are welcomed to join. If you would like to be involved in research case studies and collaborate with like-minded people, please see the IFoA volunteer vacancies page for how to apply.

For the latest news, workstream case studies and data science resources, please visit our Github page.

Objectives of Data Science Research workstream:

1. Identify interesting actuarial data science research problems and data sets
2. Conduct data science research and case studies, including but not limited to pricing, underwriting, mortality and morbidity, fraud, customer behaviour and automation.
3. Collate Data Science use cases and case studies and make them accessible

The IFoA Data Science Research Section contains multiple workstreams:

1. Natural Language Processing (NLP)
2. Supervised Learning
3. Federated Learning
4. Explainable AI and Model Interpretability. 
5. Visualisation
6. Unsupervised Learning
7. Autoencoder

Outputs

• Insurance: Collaboration without compromise, The Actuary Magazine, Mar 2021
• Keyboard Worriers, The Actuary magazine, Feb 2021
• IFoA Asia Conference, Sep 2020
• Twitter Sentiment Analysis, Jul 2020
• Data Science Event Day, Mar 2020