Progressive information technologies that have resulted from the evolution of location-based services (LBS) have significantly improved people's urban lives. Location-based social networks (LBSN) provide platforms for users to check-in and share their current locations, thoughts, experiences, and reviews of points of interest (POIs) with anyone. This enormous amount of heterogeneous data in LBSNs has enabled the development of POI recommendations. It has sparked many efforts in the research community to develop accurate POI recommendation systems in various scenarios such as mobile, automobile, and business applications. As we focus on automotive scenarios and modern and recent driver information systems, the driver has a large volume of data, such as digital broadcast information, global positioning system (GPS) information. and vehicle application information. If all of this data is transmitted to the driver unprocessed, information overload becomes a significant problem. As such, the POI recommendation service is ideally suited for mobility applications. Say no to plagiarism. Get a tailor-made essay on “Why Violent Video Games Should Not Be Banned”? Get an original essay For example, it can reduce the risk of road accidents by avoiding typing in a long place name when users search for places to go. Therefore, it not only makes it easy for users to discover new locations, but it also helps users get relevant POIs without wasting a lot of time on searching, especially when they are in a new area. In previous research, common problems encountered in POI recommender systems are cold start and data sparsity. The cold start issue is caused by limited activity history of users and locations in the system. For a new user or location, the recommendation model does not have enough information to give useful recommendations. Due to the rapid growth in the number of new users on LBSNs, the problem is getting even worse. Similarly, data sparsity is because the total data in the recommendation model is not enough to process and recognize related users/items. Therefore, hybrid approaches and novel methods taking into account different types of recommendation models are necessary...