Abstract
There are approaches that successfully recognize activities of daily living by using a trained classifier on feature vectors created from binary sensor data. Although these approaches have been successful, there are still open issues such as the evaluation of multiple temporal windows, ensembles of classifiers or unbalanced classes which need to be addressed in order to improve the performance of the real-time activity recognition process. In this paper, we present a methodology for Real-Time Activity Recognition based on the diverse fields of Machine Learning, including Fuzzy Logic and Recurrent Neural Networks. The methodology uses a long-term and short-term representation of binary-sensor activations based on Fuzzy Temporal Windows. The paper proposes an ensemble of activity-based classifiers for the purposes of balanced training, where each classifier in the ensemble is a Long Short-Term Memory. The approach was evaluated using two binary-sensor datasets of daily living activities and benchmarked against previous approaches based on the combination of sensor activation features.
Original language | English |
---|---|
Pages (from-to) | 441-453 |
Number of pages | 13 |
Journal | Expert Systems with Applications |
Volume | 114 |
Early online date | 1 Aug 2018 |
DOIs | |
Publication status | Published online - 1 Aug 2018 |
Keywords
- Activity recognition
- Fuzzy temporal windows
- Long short-term memory
- Unbalanced data
- Ensemble architectures
Fingerprint
Dive into the research topics of 'Ensemble classifier of long short-term memory with fuzzy temporal windows on binary sensors for activity recognition'. Together they form a unique fingerprint.Profiles
-
Christopher Nugent
- School of Computing - Professor of Biomedical Engineering
- Faculty Of Computing, Eng. & Built Env. - Full Professor
Person: Academic
-
Shuai Zhang
- School of Computing - Senior Lecturer in Computing Science
- Faculty Of Computing, Eng. & Built Env. - Senior Lecturer
Person: Academic