The current developments in the fields of networking and sensor technology enable more and more number of devices of the real world to be an integral part of Internet of Things (IoT). This leads to the generation of huge volumes of data on the scale sufficient enough to qualify as an instance of big data. This big data includes various types of hidden information that are useful for different kinds of services and applications. In this paper, an enhanced big data based framework for Ambient Assisted Living (AAL) system is presented. Where the prediction accuracy of the existing state-of-art work is enhanced by using map-reduce genetic algorithm. The results show that the proposed framework is more accurate when compared with the existing works.