Location-aware recommendation is considered as one of human behavior cognitive analyses in the world of human-machine-environment system. The development of 5G technology and ubiquitous mobile devices has led to the emergence of a new online platform, location-based social networks (LBSNs), which allows users to share their locations. The essential feature of LBSNs is to provide users with location recommendations that help them explore new places and also to make LBSNs more prevalent to users. Most of the existing research is focusing on the introduction of new features and how these new features affect the check-in behaviors of the users. In addition, the dependencies between each feature and the probability of a user visiting the site is always a principle to follow. However, a user’s decision could be determined by considering several features at the same time. When a full model is applied by considering all the features, an overfitting problem could be occurred owing to the lack of sufficient data for each individual user. In this article, an intermediate solution was proposed to address all of these problems by fragmenting the model into several partial models, where each partial model is responsible for a few features. An additive strategy was also implemented to support the development of personalized partial models. Furthermore, a partition-based approach was introduced to explore the hidden patterns from the geographically clustered check-in data. The performance of the approaches has been evaluated by using the data sets from Foursquare and it demonstrates that the proposed approach outperforms the state-of-the-art approaches.

Through the availability of location-acquisition devices, huge volumes of spatio-temporal data recording the movement of people is provided. Discovery of the group of people who travel together can provide valuable knowledge to a variety of critical applications. Existing studies on this topic mainly focus on the movement of vehicles or animals with forcing the group members to stay always connected. However, the movement of people is different; people might belong to the same main group while they contribute in various sub-groups during their movement. In this paper, we propose a group pattern called loose travelling companion pattern (LTCP), which allows the members of a group to contribute to various sub-groups as long as the community of members does not change during the movement and all of the members stay connected for a few time-slots. In addition, we propose weakly continuous loose travelling companion pattern (WCLTCP) to relax the continuous time constraint in LTCP. Finally, three algorithms have been developed to discover the proposed group patterns: (i) straightforward approach, (ii) smart-and-fast method, and (iii) and opportunistic algorithm. Through the extensive experimental evaluation on both real and experimental datasets, the efficiency and effectiveness of the proposed group discovery approaches are proven.