Recently, social media enhanced learning has become more and more popular. It is featured as learning through interaction and collaboration in a community or across a social network, which can be considered as a kind of social learning. In this study, we integrate SNS (such as twitter) into the web-based learning process and further delve into the discovery of potential information from the reorganized stream data. We propose a Dynamical Socialized User Networking (DSUN) model which represents users' profiling and dynamical relationship by a set of measures. Finally, we show an application scenario of the DSUN model to assist the learning process and enhance the learning efficiency in web-based environments.