Data Mining Using Human Motions for Intelligent Systems

In recent years, robot systems have come to play an important assisting role in providing information to human beings. The great progress made in Internet technology is also greatly beneficial as a means of collecting information. However, amassing too much information also gives rise to the problem about how to offer the most suitable data. In this paper, the authors present a system that can use users’ intentions to provide information to them. We focus on applying a series of “Time logs” constructed by using explicit human motions and motion time to provide base data for data mining, using the datamining to ascertain human intentions from the Time logs series, and then offering information to people. To achieve our research targets, we constructed our systembased on the Kukanchi concept, which allows every part of a system to provide and receive data. Moreover, our system uses cameras to collect human motion data without any need to place devices on users’ bodies. Our work verifies the possibility of using the Kukanchi concept to construct a system and the ability of such a system to apply human motions as mining data without disturbing users in any way. It also demonstrates the possibility of using a series of Time logs to detect human intentions so as to offer useful data to users.