Abstract
Human activity recognition (HAR) using wearable sensors is an increasingly active research topic in machine learning, aided in part by the ready availability of detailed motion capture data from smartphones, fitness trackers, and smartwatches. The goal of HAR is to use such devices to assist users in their daily lives in application areas such as healthcare, physical therapy, and fitness. One of the main challenges for HAR, particularly when using supervised learning methods, is obtaining balanced data for algorithm optimisation and testing. As people perform some activities more than others (e.g., walk more than run), HAR datasets are typically imbalanced. The lack of dataset representation from minority classes hinders the ability of HAR classifiers to sufficiently capture new instances of those activities. We introduce three novel hybrid sampling strategies to generate more diverse synthetic samples to overcome the class imbalance problem. The first strategy, which we call the distance-based method (DBM), combines Synthetic Minority Oversampling Techniques (SMOTE) with Random_SMOTE, both of which are built around the k-nearest neighbors (KNN). The second technique, referred to as the noise detection-based method (NDBM), combines SMOTE Tomek links (SMOTE_Tomeklinks) and the modified synthetic minority oversampling technique (MSMOTE). The third approach, which we call the cluster-based method (CBM), combines Cluster-Based Synthetic Oversampling (CBSO) and Proximity Weighted Synthetic Oversampling Technique (ProWSyn). We compare the performance of the proposed hybrid methods to the individual constituent methods and baseline using accelerometer data from three commonly used benchmark datasets. We show that DBM, NDBM, and CBM reduce the impact of class imbalance and enhance F1 scores by a range of 9–20 percentage point compared to their constituent sampling methods. CBM performs significantly better than the others under a Friedman test, however, DBM has lower computational requirements.
Highlights
Human activity recognition (HAR) using body-worn or wearable sensors is an active research topic in mobile and ubiquitous computing [1]
cluster-based method (CBM) performs significantly better than the others under a Friedman test, distance-based method (DBM) has lower computational requirements
We show that the sampling methods are only useful to improve the performance of the Multilayer perceptron (MLP) compared to the other classifiers for imbalanced human activity data
Summary
Human activity recognition (HAR) using body-worn or wearable sensors is an active research topic in mobile and ubiquitous computing [1]. Activity recognition is a useful tool because it provides information on an individual’s behaviour that enables computing systems to monitor and to analyse and assist with a range of day-to-day tasks [2,3]. Most HAR studies adopt a supervised learning approach [4]. Supervised learning typically requires immense amounts of labelled sensor data in order to train [2]. For such models to work well, the data are ideally recorded from a variety of real-word situations. A diversity of sensor modalities and placements can help improve recognition performance [5,6]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.