A Human-Centered Software Mechanism for the Assessment of Automatic Classification of Possible Depressive Profiles

• Grant number: 22/16364-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: March 01, 2023
• End date: February 29, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Vânia Paula de Almeida Neris
• Grantee: Vanessa de Cássia Alves
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

Major depression is a serious illness that affects more than 15% of the Brazilian population. Among college students, the prevalence is even higher. Researchers from the Amive project, funded by FAPESP, are training a computational model to identify university students with a possible depressive profile. The model considers data from mobile sensors and online social networks. The outputs of the classification model influence the delivery of interventions designed within the scope of the Amive project. Based on the classification, a chatbot will start conversations about determinants of mental suffering in the university context, offering health education content and pointing out support techniques and places. Thus, it is important to have quality in the classification and consider the user's opinion before delivering the interventions. Current solutions that consider users to improve automatic classification have explored dedicated user interfaces. This scientific initiation project aims to specify a software mechanism, which is instantiated in a chatbot and, therefore, dialogues to evaluate the classification made by the computational model. After the evaluation, the mechanism must guide the subsequent actions of the software infrastructure for more accurate delivery of interventions, in line with the Human-Centered Artificial Intelligence approach. As research methods, a Focus Group with specialists in mental health will be adopted to propose the mechanism, and data collected from students in everyday university situations, in a larger Clinical Study, will be used to evaluate the mechanism.