A comparative study of pattern recognition classifiers to predict physical activities using smartphones and wearable body sensors

This paper presents a wireless body area network platform that performs physical activities recognition using accelerometers, biosignals and smartphones. Multiple classifiers and sensor combinations were examined to identify the classifier with the best recognition performance for the static and dynamic activities. The Functional Trees classifier proved to provide the best results among the classifiers evaluated (Naive Bayes, Bayesian Networks, Support Vector Machines and Decision Trees [C4.5, Random Forest]) and was used to train the model which was implemented for the real time activity recognition on the smartphone. The identified patterns of daily physical activities were used to examine conformance with medical advice, regarding physical activity guidelines. An algorithm based on Skip Chain Conditional Random Fields, received as inputs the recognized activities and data retrieved from the GPS receiver of the smartphone to develop dynamic daily patterns that enhance prediction results. The presented platform can be extended to be used in the prevention of short-term complications of metabolic diseases such as diabetes.