A Self-Adaptive Context-Aware Model for Mobile Commerce

摘要

Use of mobile devices for the online shopping is growing ever. This paper addresses the problem of querying the contents relevant to the current context of the mobile node. We present a context-aware model that can incrementally learn the user preferences and location-based content retrieval for the purpose of one-to-one marking strategy. The model is based on Monte-Carlo sampling and tree induction method. Monte-Carlo sampling is used to construct the synopsis structure while tree induction is used to predict the user preferences in the current context. The model is evaluated using two benchmark datasets for offline testing and an application is developed to test the model online. The results show an obvious advantage of using the Monte-Carlo based tree induction method as compare to its state-of-the-art rivals.