The analysis of human location histories is currently getting an increasing attention, due to the widespread usage of geopositioning technologies such as the GPS, and as well as online location-based services that allow users to share this information. Tasks such as the prediction of person’s movement can be addressed through the usage of this data, in order for offering support for more privileged applications, such as adaptive mobile services through proactive context-based functions. Here we aim to develop a simple and effective scheme to predict when the user will leave the current location and where he will move to future position. This paper presents a hybrid method for predicting human mobility on the basis of Mobility Markov Chain Models (MMCs). The proposed approach clusters location histories according to their characteristics, and later trains the MMC model based on mobility history to obtain transition matrix. The usage of MMC allows us to take information of location characteristics as parameters, and also to account for the effects of each individual’s previous actions. The proposed system is a mobility prediction with adaptive duty cycling approach in reducing energy consumption, with a mobility model called Mobility Markov Chain (MMC) for predicting the future location. We report a series of experiments with a real-world location history dataset and from the LifeMap dataset, showing that the prediction accuracy is in the range of 65 to 85 percent.