A knowledge-light approach to personalised and open-ended human activity recognition
- Submitting institution
-
Robert Gordon University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- Wiratunga_3
- Type
- D - Journal article
- DOI
-
10.1016/j.knosys.2020.105651
- Title of journal
- Knowledge-Based Systems
- Article number
- 105651
- First page
- 105651
- Volume
- 192
- Issue
- -
- ISSN
- 0950-7051
- Open access status
- Compliant
- Month of publication
- February
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The work is part of the Horizon SelfBACK project (No:689043, 4.92million EUR) where authors lead the ML and digital health adherence monitoring work-package (http://www.selfback.eu/consortium.html). We subsequently provided the MEx dataset, publicly accessible from https://archive.ics.uci.edu/ml/datasets/MEx.
MEx is unique in that it is a multi-modal dataset related to physiotherapy exercise adherence monitoring with sensors. This paper led to an invited talk on “Learning to Compare with Few Data for Personalised Human Activity Recognition” (http://iccbr20.org/programme/invited-speakers) and a related publication (https://doi.org/10.1007/978-3-030-58342-2_1) was followed by a further invitation to be on the NHS 'Health and social innovation challenges for SMEs' panel -
https://www.candoinnovation.scot/people/nirmalie-wiratunga/.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
