A Comprehensive Study of Activity Recognition Using Accelerometers

• DOI: 10.20944/preprints201803.0147.v1
• Publication Date: 2018-03-21T23:15:18Z
• Authors: Niall Twomey; Tom Diethe; Xenofon Fafoutis; Atis Elsts; Ryan McConville; Peter Flach; Ian Craddock
• Keywords: /dk/atira/pure/core/keywords/digital_health; name=Digital Health; accelerometers; /dk/atira/pure/core/keywords/digital_health; name=Digital Health; name=SPHERE; Information technology; 02 engineering and technology; sensors; T58.5-58.64; 004; machine learning; acelerometers; 0202 electrical engineering, electronic engineering, information engineering; /dk/atira/pure/core/keywords/eng_sphere; artificial_intelligence_robotics; activity recognition; activities of daily living; /dk/atira/pure/core/keywords/eng_sphere; name=SPHERE

This paper serves a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. We examine research that has focused on the selection of activities, the features that are extracted from the accelerometer data, the segmentation of the time-series data, the locations of accelerometers, the selection and configuration trade-offs, the test/retest reliability, and the generalisation performance. Furthermore, we study these questions from an experimental platform and show, somewhat surprisingly, that many disparate experimental configurations yield comparable predictive performance on testing data. Our understanding of these results is that the experimental setup directly and indirectly defines a pathway for context to be delivered to the classifier, and that, in some settings, certain configurations are more optimal than alternatives. We conclude by identifying how the main results of this work can be used in practice, specifically in experimental configurations in challenging experimental conditions.