How Python Can Help Detect Early Signals in University Mental Health
summary
A project uses Python to analyze survey data from about 2,000 students (≈10% response rate) applying statistical tests (chi-square, t-test, ANOVA, Spearman) in Google Colab with pandas, NumPy, SciPy and Seaborn for EDA and reporting. It fine-tunes a Spanish BERT (Beto) for NLP to classify emotions (89% accuracy) and implements an alert system that reveals identities only for medium/high alerts, deployed with a Flutter mobile app and a web stack (React, NestJS, PostgreSQL).
AI-written summary · about
chapters (13)
- 0:00Latin America context
- 3:35Early detection challenges
- 7:53Clinical goals
- 9:17Data collection with Python
- 11:28Student risk factors
- 15:22NLP for emotions
- 17:53AI limits and human role
- 19:18Technical architecture
- 23:30Demo and features
- 27:19Q&A
- 30:51Open source availability
- 33:25Scalability and anonymity
- 35:33High school adaptation
related talks
22mLet’s make the world a better place: Python for Social Good
34mSolving Crimes with Python Using Natural Language Processing (NLP)
01h 23mBeyond Sentiment Emotion Mining with Python and machine learning
43mTowards a full stack python monitoring+analysis framework
34mFighting COVID with Python
36mPython in Medicine: analysing data from mechanical...
31mProgramming, Python, and Student-Directed Projects
43mEveryone can do Data Science in Python