Lecture: 'Tell Me What You Post and I’ll Tell You How You Are: Two Approaches for Detecting Depression Through Social Media Content'

The Internet has become a massive communication medium. Some services, such as social media, allow users to share personal information very easily, providing countless new opportunities for business intelligence, application personalization, cybersecurity, and public health monitoring.

This lecture will focus on detecting mental disorders through social media, particularly identifying users suffering from depression. Two complementary approaches developed by our research group will be presented: one centered on analyzing emotions expressed in users’ posts, and another on using and adapting large language models to globally analyze their publications. The lecture will conclude with a brief overview of other research topics in our lab, such as hate speech detection and propaganda content analysis.

About the speaker

Dr. Manuel Montes is a researcher at the Computational Sciences Coordination of the National Institute of Astrophysics, Optics, and Electronics (INAOE) in Mexico. His research focuses mainly on information retrieval, text mining, and authorship analysis, fields in which he has published around 300 articles in international journals and conferences. His work has earned him recognition as a National Researcher Level II by the SNI and a member of the Mexican Academy of Sciences.

Dr. Montes has been a visiting professor at the Polytechnic University of Valencia and the University of Alabama at Birmingham. He is also a founding member of the Mexican Academy of Computing and the Mexican Association for Natural Language Processing. In the context of the latter, he has organized several editions of the National Workshop on Language Technologies, the Mexican Workshop on Plagiarism Detection and Authorship Analysis, the Autumn School on Language Technologies, and evaluation tasks on author profiling, aggressive language analysis, fake news detection, and reporting violent events at IberLEF.