A knowledge-light approach to personalised and open-ended human activity recognition.
Wijekoon, Anjana; Wiratunga, Nirmalie; Sani, Sadiq; Cooper, Kay
Professor Nirmalie Wiratunga firstname.lastname@example.org
Professor Kay Cooper email@example.com
Human Activity Recognition (HAR) is a core component of clinical decision support systems that rely on activity monitoring for self-management of chronic conditions such as Musculoskeletal Disorders. Deployment success of such applications in part depend on their ability to adapt to individual variations in human movement and to facilitate a range of human activity classes. Research in personalised HAR aims to learn models that are sensitive to the subtle nuances in human movement whilst Open-ended HAR learns models that can recognise activity classes out of the pre-defined set available at training. Current approaches to personalised HAR impose a data collection burden on the end user; whilst Open-ended HAR algorithms are heavily reliant on intermediary-level class descriptions. Instead of these 'knowledge-intensive' HAR algorithms; in this article, we propose a 'knowledge-light' method. Specifically, we show how by using a few seconds of raw sensor data, obtained through micro-interactions with the end-user, we can effectively personalise HAR models and transfer recognition functionality to new activities with zero re-training of the model after deployment. We introduce a Personalised Open-ended HAR algorithm, MNZ, a user context aware Matching Network architecture and evaluate on 3 HAR data sources. Performance results show up to 48.9% improvement with personalisation and up to 18.3% improvement compared to the most common 'knowledge-intensive' Open-ended HAR algorithms.
WIJEKOON, A., WIRATUNGA, N., SANI, S. and COOPER, K. 2020. A knowledge-light approach to personalised and open-ended human activity recognition. Knowledge-based systems [online], 192, article ID 105651. Available from: https://doi.org/10.1016/j.knosys.2020.105651
|Journal Article Type||Article|
|Acceptance Date||Feb 9, 2020|
|Online Publication Date||Feb 13, 2020|
|Publication Date||Mar 15, 2020|
|Deposit Date||Feb 11, 2020|
|Publicly Available Date||Feb 14, 2021|
|Peer Reviewed||Peer Reviewed|
|Keywords||Human activity recognition; Personalised HAR; Open-ended HAR; Zero-shot learning; Matching networks|
This file is under embargo until Feb 14, 2021 due to copyright reasons.
Contact firstname.lastname@example.org to request a copy for personal use.
You might also like
Learning to recognise exercises for the self-management of low back pain.
Preface: case-based reasoning and deep learning.
Human activity recognition with deep metric learners.
Personalised meta-learning for human activity recognition with few-data.
Heterogeneous multi-modal sensor fusion with hybrid attention for exercise recognition.