Machine Learning Based Adaptive Context-Aware System for Smart Home Environment

Abstract

Context-awareness is the key element for building a smart home environment. The goal of a smart home is to predict the demand of home users and proactively provides the proper services by considering the user’s context information. Several methods are used in context-aware system to provide services. Machine learning based approaches are capable to make better prediction and adaptation than others. In this paper, we present machine learning based context-aware system which can provide service according to the trained model. Two effective learning algorithms: Back propagation Neural Network, and Temporal Differential (TD) class of reinforcement learning are used for prediction and adaptation respectively. This approach indicates better adaptation for context-aware service due to the low error rate.