Behavioral interval estimation model construction device, behavioral interval estimation model construction method, and behavioral interval estimation model construction program

摘要

PROBLEM TO BE SOLVED: To efficiently construct an action interval estimation model. SOLUTION: A second hidden Markov model is each of a plurality of first hidden Markov models whose state is a type of human motion, and is a state of an action determined by combining a plurality of motions. In the hidden semi-Markov model including the second hidden Markov model, the observation probabilities for each type of operation of the plurality of first hidden Markov models are learned by unsupervised learning. The learned observation probability is fixed, the input first supervised data is inflated to become the second supervised data, and the transition probability of the operation of the first hidden Markov model is the supervised learning using the second supervised data. Learn at. We construct a hidden semi-Markov model, which is a model that estimates the interval of behavior using the learned observation probability and transition probability. [Selection diagram] FIG. 15