Fig. 5From: Animal behaviour on the move: the use of auxiliary information and semi-supervision to improve behavioural inferences from Hidden Markov Models applied to GPS tracking datasets Example of a foraging trip from the dataset informed with auxiliary data Large circles indicate positions used as test states to measure accuracy, while small circles represent those included in the model as known states. (A) Positions colored by known states (yellow = resting, red = foraging, cyan = travelling). (B) Positions coloured by correct (green) or incorrect (red) classification by a HMM with no known states (0%) informed by auxiliary data. (C) Positions coloured by correct (green) or incorrect (red) classification by a HMM in which we included 75% known states informed by auxiliary data. (D) Positions coloured by the HMM probability of classification, red points have a probability < 0.9 while grey points have a probability > 0.9Back to article page