An Efficient Fuzzy Clustering Trajectories of Mobile Objects in Road Networks using Depth First Search

Most of mobile object trajectory clustering analysis to date has been focused on clustering the location points or sub-trajectories extracted from trajectory data. In this paper presents Fuzzy based Rapid Locality-Aware Trajectory Pattern Mining (FRLAT), a systematic approach to soft clustering whole trajectories of mobile objects travelling in road networks. FRLAT as a whole trajectory clustering framework has three unique features. First, we design Locality-aware Partitioned (LP) Mobile id-lists in whole trajectories. Second, we develop an Depth First Traversal (DFS) to discover interesting paths between locations in the given trajectory dataset. Third Fuzzy based Trajectory Mapping process is to better optimize the whole trajectory clustering process into multidimensional data points in a Euclidean space while preserving their relative distances in the transformed metric space.