A user customized service provider framework based on machine learning

In this paper, we propose a user customized service provider framework based on machine learning. The framework consists of mobile stations, data collector, analysis tools and service applications. As an analysis tool, we deploy machine learning techniques, in particular, support vector machine which generates learning model and precise classifiers. Moreover, K-fold cross-validation is used to achieve better accurate inference from the collected data. Then, we develop a predictor that predicts users´ behavior patterns from the information of time connections and APs. This enables to provide adaptive services customized for end-users, e.g., smart phone push notifications services.