User Behavior Recognition based on Clustering for the Smart Home

In the vision of ubiquitous computing environment, smart objects would communicate each other and provide many kinds of information about user and their surroundings in the home. This information enables smart objects to recognize user behaviors and to provide active and convenient services to the customers. However in most cases, context-aware services are available only with expert systems. The proposed method presents general context-aware system in the smart home based on a Bayesian network (BN), which enables us handling user context probabilistically. We used fuzzy c-means algorithm for clustering user positions and active time to simplify BN’s structure and to reduce BN's conditional probability table size. We used a virtual smart home with web based simulator to collect samples of user activities and their environments. This suggested algorithm was simulated with 10 fold cross-validation and evaluated to verify the performance of context-awareness. Key-Words: Behavior recognition, Context-aware, Smart home, Bayesian network, Fuzzy c-mean clustering