A framework to handle data heterogeneity contextual to medical big data

A massive growth of collaborative frameworks, where web services, internet of things, mobile applications and digitization of enterprise processes leads to generate massively heterogeneous data termed as Big Data. Handling the heterogeneity of such data by a distributed file management system based database such as H-Base having limitation to handle only structured form of the data. This paper introduces an efficient algorithm for managing data heterogeneity in three basic types i.e. 1) centralized structured data to distributed structured data, 2) unstructured to a structured format and 3) semi-structured to a structured format. The performance of proposed method is evaluated in a real-time prototype experiment and in future we planned to compare to work on a data transform method in the specific context of health care data addressing performance metrics such as memory consumption and request per write. The experimental outcomes of the proposed system show how the processing time is reduced even when we process data of large size, thereby showing the effectiveness of presented approach.