An infrastructure and approach for inferring knowledge over big data in the vehicle insurance industry

Abstract

In recent years, insurance organizations have turned their attention to tapping into massive amounts of data that are continuously produced across their IT ecosystem. Even though the concept of Big Data provides the needed infrastructure for efficient data management, especially in terms of storage and processing, the aspects of Value and Variety still remain a topic for further investigation. To this end, we propose an infrastructure that can be deployed on top of the legacy databases of insurance companies. The ultimate aim of this attempt is to provide an efficient manner to access data on-the-fly and derive new value. In our work, we propose a Property and Casualty ontology and then exploit an OBDA system in order to leverage its power.