Detecting Unseen Falls from Wearable Devices using Channel-wise Ensemble of Autoencoders

Abstract

A fall is an abnormal activity that occurs rarely, so it is hard to collect real data for falls. It is, therefore, difficult to use supervised learning methods to automatically detect falls. Another challenge in using machine learning methods to automatically detect falls is the choice of engineered features. In this paper, we propose to use an ensemble of autoencoders to extract features from different channels of wearable sensor data trained only on normal activities. We show that the traditional approach of choosing a threshold as the maximum of the reconstruction error on the training normal data is not the right way to identify unseen falls. We propose two methods for automatic tightening of reconstruction error from only the normal activities for better identification of unseen falls. We present our results on two activity recognition datasets and show the efficacy of our proposed method against traditional autoencoder models and two standard one-class classification methods.

25 pages, 6 figures, 4 Tables

References in corpus (4)

• One-Class Classification: Taxonomy of Study and Review of Techniques
• Review of Fall Detection Techniques: A Data Availability Perspective
• Detecting Falls with X-Factor Hidden Markov Models
• Modular Autoencoders for Ensemble Feature Extraction

Cited by in corpus (7)

• Deep Learning for Sensor-based Human Activity Recognition: Overview, Challenges and Opportunities
• Cross-domain Activity Recognition via Substructural Optimal Transport
• Spatio-Temporal Adversarial Learning for Detecting Unseen Falls
• Machine Learning for the Detection and Identification of Internet of Things (IoT) Devices: A Survey
• Activity2Vec: Learning ADL Embeddings from Sensor Data with a Sequence-to-Sequence Model
• A Mobile Cloud Collaboration Fall Detection System Based on Ensemble Learning
• One-Class Classification by Ensembles of Regression models -- a detailed study