Semantic lifting and reasoning on the personalised activity big data repository for healthcare research

The fast growing markets of smart health monitoring devices and mobile applications provide opportunities for common citizens to have capability for understanding and managing their own health situations. However, there are many challenges for data engineering and knowledge discovery research to enable efficient extraction of knowledge from data that is collected from heterogonous devices and applications with big volumes and velocity. This paper presents research that initially started with the EC MyHealthAvatar project and is under continual improvement following the project's completion. The major contribution of the work is a comprehensive big data and semantic knowledge discovery framework which integrates data from varied data resources. The framework applies hybrid database architecture of NoSQL and RDF repositories with introductions for semantic oriented data mining and knowledge lifting algorithms. The activity stream data is collected through Kafka's big data processing component. The motivation of the research is to enhance the knowledge management, discovery capabilities and efficiency to support further accurate health risk analysis and lifestyle summarisation.