UNRAVELING MARKOV PROCESSES IN MOVEMENT PATTERNS OF INDICATOR SPECIES IN RESPONSE TO CHEMICAL STRESSORS
TUYEN VAN NGUYEN, Yuedan Liu, Il-Hyo Jung, Tae‐Soo Chon, Sang‐Hee Lee · Modern Physics Letters B · 2011
Revealing biological responses of organisms in responding to environmental stressors is the critical issue in contemporary ecological sciences. Markov processes in behavioral data were unraveled by utilizing the hidden Markov model (HMM). Individual organisms of daphnia (Daphnia magna) and zebrafish (Danio rerio) were exposed to diazinon at low concentrations. The transition probability matrix (TPM) and the emission probability matrix (EPM) were accordingly estimated by training with the HMM and were verified before and after the treatments with 10-6 tolerance in 103 iterations. Structured property in behavioral changes was accordingly revealed to characterize dynamic processes in movement patterns. Parameters and sequences produced through the HMM training could be a suitable means of monitoring toxic chemicals in environment.