This paper presents three behavior labeling algorithms based on supervised learning using accumulated pyroelectric sensor data in the living space. We summarize features of each algorithm to use them in combination matched to usage of the livelihood support application. They are (a)labeling algorithms based on switching model around a behavioral change-point, (b)one based on time attribution of "on-off" data, and (c)one based on Hidden Markov Models. We show the behavior labeling results of three algorithms for one month data under the same conditions. We summarize features on the basis of these results.