Learning Daily Activity Recognition Model with Sharing Training Data and Semi-Supervised Learning

Abstract

We propose a new daily activity recognition method that can learn an activity classification model with small quantities of training data by sharing training data among different activity classes. Many existing activity recognition studies employ a supervised machine learning approach and thus require an end user's labeled training data, this approach places a large burden on the user. In this study, we assume that a user wears sensors (accelerometers) on several parts of the body such as the wrist, waist, and thigh, and we attempt to share sensor data obtained from only selected accelerometers (e.g., only waist and thigh sensors) among two different activity classes. For further reducing the burden on the user, we also adopt a semi-supervised approach.