Abstract
There has been an increase in the need to have a weight management system that prevents adverse health conditions which can in the future lead to various
cardiovascular diseases. Several types of research were made in attempting to understand and better manage body-weight gain and obesity.
This study focuses on a data-driven approach to identify patterns in profiles with body-weight change in a dietary intervention program using machine learning algorithms. The proposed line of investigation would analyse these patient’s profile at the entry of dietary intervention program and for some, on a weekly basis. These attributes would serve as inputs into machine learning algorithms.
From the unsupervised learning perspective, the paper seeks to address the first stage in applying machine learning algorithms to weight management data. The specific aim here is to identify the thresholds for weight loss categories which
are required for supervised learning.
Original language | English |
---|---|
Title of host publication | The 10th IEEE International Conference on Dependable Systems, Services and Technologies. DESSERT'2019 : Conference proceedings |
Number of pages | 8 |
Publisher | IEEE |
Publication date | Jun 2019 |
Pages | 94-101 |
Article number | 8770032 |
ISBN (Electronic) | 978-1-7281-1733-1 |
DOIs | |
Publication status | Published - Jun 2019 |
Event | IEEE International Conference on Dependable Systems, Services and Technologies: DESSERT'2019 - Leeds Beckett University, Leeds, United Kingdom Duration: 5 Jun 2019 → 7 Jun 2019 Conference number: 10 http://dessert.ieee.org.ua/dessert-2019/program/ |
Conference
Conference | IEEE International Conference on Dependable Systems, Services and Technologies |
---|---|
Number | 10 |
Location | Leeds Beckett University |
Country/Territory | United Kingdom |
City | Leeds |
Period | 05/06/2019 → 07/06/2019 |
Internet address |
Keywords
- Faculty of Science
- Weight management
- Weight loss categorisation
- Unsupervised learning
- Data clustering
- Smart health management