The purpose of the study was to predict the likelihood of a claim being part of an accident. We developed the algorithm that could leverage features, both from the claims and potentially beyond the claims.
We inserted an algorithm into our client’s existing ETL process and provided a single, key metric: likelihood of this claim being part of an accident. In addition, we demonstrated the expected claims whose decisions changed between actual historical decision and modeled decision. We helped our client estimate profitability lift.