Pattern classification of movement behaviors of indicator specimens using permutation entropy after treatments with Diazinon

Movement behaviors of Daphnia magna and Medaka (Oryzias latipes) were computationally analyzed before and after the treatments of diazinon. During the observation period, individual test specimens in 2-D were continuously recorded in an aquarium at 0.25 second intervals. By ordering positional data, permutation entropy was measured to extract information residing in the time series data of the movement tracks. By checking the successive or equal shift in positional data, permutation entropy accordingly revealed various states of response behaviors of test specimens. The obtained data for permutation entropy were additionally compared between Daphnia magna and Medaka. Efficiency of permutation entropy was further discussed regarding analysis of abnormal behaviors of indicator specimens and monitoring toxic chemicals in aquatic environment.