Review of Trajectories Similarity Measures in Mining Algorithms

Trajectory similarity can be defined as the cost of transforming one trajectory into another based on certain similarity function. It is the core of numerous mining tasks such as clustering, classification, outlier detection, and indexing. Various approaches have been suggested to measure the similarity between pair of trajectories based on their geometric properties, the overlapping between their segments, the confined area between them, and semantic concept. This study aims to highlight and evaluate these approaches in term of their computational cost, usage memory, accuracy, and the amount of data, which is required to process in advance in order to determine its suitability to static or stream mining applications. The evaluation results concludes that the stream mining applications support similarity methods which have low computational cost and memory, single scan on data, and free of mathematical complexity due to high speed generation of data.