All clear: How Shapley values make opaque models more transparents
Originally published by The Actuary, March 2022. © The Institute and Faculty of Actuaries. Click here to read the original article.
Posts by Category
Blog
Applying Analytics in Embedded Insurance
Introduction The past decade has seen the rise of technology penetration in our everyday lives, from ride-hailing, food delivery, e-commerce, digital bank...
Splitting dataset for federated learning
In this post, we would explore how to split a dataset for the purpose of federated learning. Federated learning is applicable when there are multiple indepen...
Supervised learning techniques in claims frequency modelling
Originally published by The Actuary, July 2021. © The Institute and Faculty of Actuaries. Click here to read the original article.
From raw to processed tweets
This is part of a series of posts that aim to explain in greater detail the NLP pipeline supporting the twitter sentiment analysis performed here. Step 1:...
Insurance: Collaboration without compromise
Originally published by The Actuary, March 2021. © The Institute and Faculty of Actuaries. Click here to read the original article. Małgorzata Śmietanka ...
All together now: modelling claims using federated learning
Originally published by The Actuary, September 2021. © The Institute and Faculty of Actuaries. Click here to read the original article.
Insurance and social media: Keyboard warriors
Originally published by The Actuary, February 2021. © The Institute and Faculty of Actuaries. Click here to read the original article. John Ng and Melani...
Pieces of the technology puzzle
Originally published by The Actuary, July 2020. © The Institute and Faculty of Actuaries. Click here to read the original article. Małgorzata Śmietanka a...
Tutorial
Presentation
Differential Privacy and Fairness in Automated Decision Making?
Orginally presented in the IFoA Research Section Knowledge Sharing Session, February 2022 Link to video Link to slides
Federated Learning: Collaboration with no compromise
Orginally presented in the Data Science Research Section monthly sharing session, September 2021 Download slides
NLP and ML in Social Media Opinion Mining
Orginally presented at the EAA e-Conference on Data Science & Data Ethics, June 2021 Link to event Download slides
Modular Framework of Machine Learning Pipeline
Orginally presented in the IFoA Data Science Webinar Series; and at the IFoA Asia Conference 2020 September 2022 Link to event Link to IFoA Virtual Learning ...
Twitter Sentiment Analysis: What does Social Media tell us about coronavirus concerns in the UK?
Orginally presented in the Data Science Research Section monthly sharing session, July 2020 Link to event Link to slides Download slides
Data Visualisation in Machine Learning
Orginally presented in the IFoA Data Science Seminar, March 2020 Link to event Download slides
Research Papers
Classification of climate-related insurance claims using gradient boosting
Originally published by The Instituto de Actuarios Espanioles, December 2022. © Instituto de Actuarios Espanioles. Link to original post. Read online ful...
EM Estimation for the Bivariate Mixed Exponential Regression Model
Orginally published in MDPI, May 2022 Link to journal post Download article
The multivariate mixed Negative Binomial regression model with an application to insurance a posteriori ratemaking
Orginally published in the Insurance: Mathematics and Economics Journal, November 2021 Link to journal post Download article Abstract This paper is conc...
A first-order binomial-mixed Poisson integer-valued autoregressive model with serially dependent innovations
Orginally published in the Journal of Applied Statistics, 1 November 2021 Link to journal post Download article Abstract Motivated by the extended Poisso...
Bivariate Mixed Poisson Regression Models with Varying Dispersion
Orginally published in the North American Actuarial Journal, 30 October 2021 Link to journal post Download article Abstract The main purpose of this artic...
Federated Learning for Privacy-preserving data access
Orginally published in SSRN, September 2020 Link to journal post Download article
Dashboards
Back to Top ↑Publications
Application of Federated Learning in Predictive Maintenance to Predict Remaining Useful Life
Submitted in partial fulfillment of the requirements for the MSc degree in Computing Science of Imperial College London, September 2021 Link to document