DATA PROCESSING APPARATUS, DATA PROCESSING METHOD, AND PROGRAM

摘要

PROBLEM TO BE SOLVED: To stochastically predict a plurality of probabilities of a user's activity state at desired future time.

SOLUTION: A sensor input part 11 obtains time-series data from a wearable sensor. A user activity model learning part 13 learns an activity model expressing a user's activity state from the obtained time-series data as a stochastic state transition mode and supplies a parameter obtained from the result of the learning to a user model parameter storage part 14. A user activity recognition part 15 recognizes a present user's activity state by using the user's activity model obtained by the learning. A user activity prediction part 16 predicts a user's activity state after the lapse of predetermined time from the present. For instance, this invention can be applied to a data processing apparatus for predicting user's behavior and allowing a control target device to perform processing corresponding to the predicted behavior.

COPYRIGHT: (C)2011,JPO&INPIT