An efficient cluster-based cab recommender system (CBCRS) provides solo cab drivers with recommendations about the next pickup location having high passenger finding potential at the shortest distance. To recommend the cab drivers with the next passenger location, it becomes imperative to cluster the global positioning system (GPS) coordinates of various pick-up locations of the geographic region as that of the cab. Clustering is the unsupervised data science that groups similar objects into a cluster. Therefore, the objectives of the research paper are fourfold: Firstly, the research paper identifies various clustering techniques to cluster GPS coordinates. Secondly, to design and develop an efficient algorithm to cluster GPS coordinates for CBCRS. Thirdly, the research paper evaluates the proposed algorithm using standard datasets over silhouette coefficient and Calinski-Harabasz index. Finally, the paper concludes and analyses the results of the proposed algorithm to find out the most optimal clustering technique for clustering GPS coordinates assisting cab recommender system.