Use of Machine Learning to Mine User-Generated Content From Mobile Health Apps for Weight Loss to Assess Factors Correlated With User Satisfaction

Authors:
Publisher website
Google Scholar
Abstract
The prevalence of obesity or overweight has been increasing yearly, and there has been rapid development of effective, efficient, and low-cost mobile health (mHealth) apps for managing obesity. In 2016, more than 900 million adults worldwide were overweight, and more than 650 million adults were obese.1 In 2019, 34.3% of Chinese adults were overweight and 16.4% were obese.2 Approximately 250 000 mHealth apps were available in 2016,3 of which more than half involved weight, diet, and exercise.4 Usability, feasibility, and acceptability of mHealth apps for management of obesity have been shown in the experimental context.5,6 One study showed that after a 12-week diet and exercise intervention, 30 individuals who used a weight loss app lost 5% of their body weight.7 In another study, after a 6-month diet intervention, the mHealth app group had a higher follow-up patient retention rate (100% vs 37%) and lower treatment costs ($855.15 vs $1428.00) than the weight management group that did not use an mHealth app.8

This publication has 31 references indexed in Scilit: