Towards Clustering of Mobile and Smartwatch Accelerometer Data for Physical Activity Recognition
Journal Title: Informatics - Year 2018, Vol 5, Issue 2
Abstract
Mobile and wearable devices now have a greater capability of sensing human activity ubiquitously and unobtrusively through advancements in miniaturization and sensing abilities. However, outstanding issues remain around the energy restrictions of these devices when processing large sets of data. This paper presents our approach that uses feature selection to refine the clustering of accelerometer data to detect physical activity. This also has a positive effect on the computational burden that is associated with processing large sets of data, as energy efficiency and resource use is decreased because less data is processed by the clustering algorithms. Raw accelerometer data, obtained from smartphones and smartwatches, have been preprocessed to extract both time and frequency domain features. Principle component analysis feature selection (PCAFS) and correlation feature selection (CFS) have been used to remove redundant features. The reduced feature sets have then been evaluated against three widely used clustering algorithms, including hierarchical clustering analysis (HCA), k-means, and density-based spatial clustering of applications with noise (DBSCAN). Using the reduced feature sets resulted in improved separability, reduced uncertainty, and improved efficiency compared with the baseline, which utilized all features. Overall, the CFS approach in conjunction with HCA produced higher Dunn Index results of 9.7001 for the phone and 5.1438 for the watch features, which is an improvement over the baseline. The results of this comparative study of feature selection and clustering, with the specific algorithms used, has not been performed previously and provides an optimistic and usable approach to recognize activities using either a smartphone or smartwatch.
Authors and Affiliations
Chelsea Dobbins and Reza Rawassizadeh
Keywords
Related Articles
- EP ID EP44131
- DOI https://doi.org/10.3390/informatics5020029
- Views 227
- Downloads 0
Interactive Spatiotemporal Analysis of Oil Spills Using Comap in North Dakota
The aim of the study is to analyze the oil spill pattern from various types of incidents and contaminants to determine the extent that incident data can be used as a baseline to prevent hazardous material releases and...
Motivation and User Engagement in Fitness Tracking: Heuristics for Mobile Healthcare Wearables
Wearable fitness trackers have gained a new level of popularity due to their ambient data gathering and analysis. This has signalled a trend toward self-efficacy and increased motivation among users of these devices. F...
Web Apps Come of Age for Molecular Sciences
Whereas server-side programs are essential to maintain databases and run data analysis pipelines and simulations, client-side web-based computing tools are also important as they allow users to access, visualize and an...
Creating a Multimodal Translation Tool and Testing Machine Translation Integration Using Touch and Voice
Commercial software tools for translation have, until now, been based on the traditional input modes of keyboard and mouse, latterly with a small amount of speech recognition input becoming popular. In order to test wh...
Understanding the EMR-Related Experiences of Pregnant Japanese Women to Redesign Antenatal Care EMR Systems
Woman-centered antenatal care necessitates Electronic Medical Record (EMR) systems that respect women’s preferences. However, women’s preferences regarding EMR systems in antenatal care remain unknown. This work aims t...