A Comprehensive Study of Activity Recognition Using Accelerometers
Journal Title: Informatics - Year 2018, Vol 5, Issue 2
Abstract
This paper serves as a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. The paper is particularly focused on long-term activity recognition in real-world settings. In these environments, data collection is not a trivial matter; thus, there are performance trade-offs between prediction accuracy, which is not the sole system objective, and keeping the maintenance overhead at minimum levels. 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.
Authors and Affiliations
Niall Twomey, Tom Diethe, Xenofon Fafoutis, Atis Elsts, Ryan McConville, Peter Flach and Ian Craddock
Keywords
Related Articles
- EP ID EP44133
- DOI https://doi.org/10.3390/informatics5020027
- Views 239
- Downloads 0
Building Realistic Mobility Models for Mobile Ad Hoc Networks
A mobile ad hoc network (MANET) is a self-configuring wireless network in which each node could act as a router, as well as a data source or sink. Its application areas include battlefields and vehicular and disaster a...
Opening up the Black Box of Sensor Processing Algorithms through New Visualizations
Vehicles and platforms with multiple sensors connect people in multiple roles with different responsibilities to scenes of interest. For many of these human–sensor systems there are a variety of algorithms that transfo...
Molecular Imaging of Bacterial Infections in vivo: The Discrimination between Infection and Inflammation
Molecular imaging by definition is the visualization of molecular and cellular processes within a given system. The modalities and reagents described here represent a diverse array spanning both pre-clinical and clinic...
Utilizing Provenance in Reusable Research Objects
Science is conducted collaboratively, often requiring the sharing of knowledge about computational experiments. When experiments include only datasets, they can be shared using Uniform Resource Identifiers (URIs) or Di...
In Search of Smartness: The EU e-Justice Challenge
At the EU level, an increasing number of resources are being invested in an attempt to provide better public services through the use of Information and Communication Technology (ICT). While new tools are being designe...