Publicação
AI Chatbots for Well-Being Adoption Drivers
| Resumo: | The global mental health sector lacks resources particularly in low- and middle-income countries. AI- chatbots are increasingly recognized as a promising resource to support mental health. However, the acceptance of AI chatbots in the field of mental health is still low. Our study aims to predict their usage, identifying the features which are more relevant for the prediction. An online survey was created in China with 400 valid responses collected. For our prediction exercise we used three Machine Learning algorithms: decision tree, logistic regression, and random forest. The accuracy of these algorithms ranged from 64-71%. Age was the most important feature to explain usage. Younger people value the compatibility of AI chatbots for well-being with their lifestyle, and older people are influenced by social factors to use them. Higher complexity can be an obstacle for AI chatbots for well-being use. |
|---|---|
| Autores principais: | Tavares, Jorge |
| Outros Autores: | Yang, Yanrong; Oliveira, Tiago |
| Assunto: | Artificial Intelligence Mental Health AI chatbots Machine Learning Technology Adoption General Computer Science SDG 3 - Good Health and Well-being |
| Ano: | 2026 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | The global mental health sector lacks resources particularly in low- and middle-income countries. AI- chatbots are increasingly recognized as a promising resource to support mental health. However, the acceptance of AI chatbots in the field of mental health is still low. Our study aims to predict their usage, identifying the features which are more relevant for the prediction. An online survey was created in China with 400 valid responses collected. For our prediction exercise we used three Machine Learning algorithms: decision tree, logistic regression, and random forest. The accuracy of these algorithms ranged from 64-71%. Age was the most important feature to explain usage. Younger people value the compatibility of AI chatbots for well-being with their lifestyle, and older people are influenced by social factors to use them. Higher complexity can be an obstacle for AI chatbots for well-being use. |
|---|