Predicting Branch Visits and Credit Card Up-selling using Temporal Banking Data

TL;DR

This paper develops feature extraction techniques from temporal banking data to predict branch visits and credit card up-selling, achieving competitive results in a data science challenge.

Contribution

It introduces a novel feature extraction approach for temporal banking data to improve prediction of user behavior in banking applications.

Findings

Ranked 4th in the ECML/PKDD 2016 challenge for branch visit prediction.

Achieved an AUC of 0.7056 for credit card up-selling prediction.

Demonstrated the effectiveness of feature engineering on temporal data.

Abstract

There is an abundance of temporal and non-temporal data in banking (and other industries), but such temporal activity data can not be used directly with classical machine learning models. In this work, we perform extensive feature extraction from the temporal user activity data in an attempt to predict user visits to different branches and credit card up-selling utilizing user information and the corresponding activity data, as part of \emph{ECML/PKDD Discovery Challenge 2016 on Bank Card Usage Analysis}. Our solution ranked \nth{4} for \emph{Task 1} and achieved an AUC of \textbf{$0.7056$} for \emph{Task 2} on public leaderboard.