A tale of two outcomes: Data Science methods for account management when things go right and when they go wrong

Two methods for managing financial accounts using different data science methods.

  • 12.10.2022, 12:00 - 13:00

A seminar by Dr. Cristián Bravo, Associate Professor and Canada Research Chair in Banking and Insurance Analytics at the University of Western Ontario. 

Reykjavik University
October 12
12PM-1PM.
Room M208

In this talk I will present two methods for managing financial accounts (credit cards and outstanding accounts) using different data science methods. For when things go right, I will showcase a reinforcement learning model that automatically identifies customers that are candidates for a credit card limit increase. The model uses a combination of a three-stage predictive model using XGBoosting and regression models as a basis, which are then fed into a double Q-learning and SARSA reinforcement learning method that optimizes the management of credit limits. The model is benchmark against common practices and we show how significant monetary and managerial improvements can be made when applying these techniques. For when things go wrong, I will present a decision support tool suitable for assessing intervention effectiveness at a customer level by leveraging two techniques. 

Firstly, we design and apply deep sequence models capable of learning representations from complex and rich customer behavioural data. Secondly, we utilise Uplift Modelling, a framework used to estimate individual-level treatment effects from observational data, to identify the optimal strategy in order to maximize the profit of the intervention done on the customers. Our results produce Qini scores ranging up to 35%, with deep sequence variants showing notable performance improvements compared against a set of baseline models. These two models showcase how modern data science methods are not only impressive in terms of their effectiveness, but also bring useful insights and make practice more efficient. 

Bio: Dr. Cristián Bravo is Associate Professor and Canada Research Chair in Banking and Insurance Analytics at the University of Western Ontario, Canada. Previously he served as Associate Professor of Business Analytics at the Department of Decision Analytics and Risk, University of Southampton, Research Fellow at KU Leuven, Belgium; and as Research Director at the Finance Centre, Universidad de Chile. His research focuses on the development and application of data science methodologies in the context of credit risk analytics, in areas such as deep learning, text analytics, image processing, and social network analysis. He has over 70 publications in high-impact journals and conferences in operational research and computer science. He also serves as editorial board member in Applied Soft Computing and the Journal of Business Analytics. He is the co-author of the book “Profit Driven Business Analytics”, with editions in English and Chinese.



Vinsamlegast athugið að á viðburðum Háskólans í Reykjavík (HR) eru teknar ljósmyndir og myndbönd sem notuð eru í markaðsstarfi HR. Hægt er að nálgast frekari upplýsingar á ru.is eða með því að senda tölvupóst á netfangið: personuvernd@ru.is
//
Please note that at events hosted at Reykjavik University (RU), photographs and videos are taken which might be used for RU marketing purposes. Read more about this on our ru.is or send an e-mail: personuvernd@ru.is