RESUMEN
Lifestyle-related diseases are a major problem all over the world although exercising can prevent them. Therefore, it is necessary to encourage users to exercise regularly and to support their exercises. The purpose of this study is to investigate whether the estimation of behavior change stages can be predicted from the gait information obtained from wearable devices, and whether message interventions created based on the behavior change stages are effective in improving the amount of walking. As for the estimation of the behavior change stages, we investigated whether the behavior change stages could be correctly estimated compared with the ones obtained from the questionnaire. As for the effect of the message, we compared the period of no intervention with that of intervention to examine whether there was any change in the amount of walking. As a result of the experiment, we could not properly estimate the behavior change stage of users, but we found that the message intervention improved the amount of walking for many subjects. This suggests that further research is needed to estimate the stage of behavior change. However, message intervention is confirmed as an effective means to improve walking volume.