Fast Similarity Search for Moving Object Trajectories Using Distributed Mining Algorithm
Object tracking applications focuses on finding the movement patterns of a single object or all objects. A mining algorithm called distributed mining algorithm is proposed to identify a group of objects with similar movement patterns. This information is important in some biological research domains, such as the study of animals, social behavior and wildlife migration. The proposed distributed mining algorithm comprises of two phases namely local mining phase and cluster ensemble phase. In local mining phase, the algorithm finds the movement patterns based on local trajectories. Based on the derived patterns from local mining phase, a new similarity measure is proposed to compute the similarities between the moving objects and also identifies the local group relationships.