Enhancing Claim Classification with Feature Extraction from Anomaly-Detection-Derived Routine and Peculiarity Profiles

TL;DR

This paper introduces a novel method using anomaly detection to create routine and peculiarity profiles from vehicle trip data, enhancing claim classification accuracy in usage-based insurance.

Contribution

It develops a new feature extraction approach from anomaly profiles that improves vehicle claim classification performance.

Findings

Features from peculiarity profiles improve classification accuracy

Anomaly scores effectively characterize vehicle trip behaviors

Method demonstrates practical utility with real vehicle data

Abstract

Usage-based insurance is becoming the new standard in vehicle insurance; it is therefore relevant to find efficient ways of using insureds' driving data. Applying anomaly detection to vehicles' trip summaries, we develop a method allowing to derive a "routine" and a "peculiarity" anomaly profile for each vehicle. To this end, anomaly detection algorithms are used to compute a routine and a peculiarity anomaly score for each trip a vehicle makes. The former measures the anomaly degree of the trip compared to the other trips made by the concerned vehicle, while the latter measures its anomaly degree compared to trips made by any vehicle. The resulting anomaly scores vectors are used as routine and peculiarity profiles. Features are then extracted from these profiles, for which we investigate the predictive power in the claim classification framework. Using real data, we find that features extracted from the vehicles' peculiarity profile improve classification.