Mobile social networks as emerging social communication platforms have attracted great attention recently, and their mobile applications have been developed and implemented pervasively. In mobile social networking applications, profile matching acts as a critical initial step to help users, especially strangers, initialize conversation with each other in a distributed manner. Introduced a distributed mobile communication system, called E-SmallTalker, which facilitates social networking in physical proximity. ESmallTalker automatically discovers and suggests common topics between users for easy conversation. E-healthcare cases by proposing a symptom matching scheme for mobile health social networks. They considered that such matching scheme is valuable to the patients who have the same symptom to exchange their experiences, mutual support, and inspiration with each other. Autoregressive model (AR) is a classic tool for understanding and predicting a time series data. It estimates the current term zk of the series by a linear weighted sum of previous p terms (i.e., observations) in the series. The model order p is generally much smaller than the length of the series. AR is often combined with Moving-Average model (MA) to obtain complex ARMA model for generally improved accuracy. While AR depends on the previous terms of a time series data, MA describes the current value of the series using a linear weighted sum of white Gaussian noise or random shocks of its prior q values. As a straightforward combination of AR and MA, ARMA model is notated as ARMA.