MO850 - Tópicos Avançados em Ciência da Computação I - 1S/2026

This repository contains course materials for MO850 – Tópicos Avançados em Ciência da Computação I (IC - UniCamp).

Introduction

The course introduces computational methods for analyzing natural language with a focus on misinformation and psychological constructs. It combines theoretical foundations from cognition and communication with practical approaches to corpus construction, semantic representation, topic modeling, and the use of large language models (LLMs). Ethical and methodological implications of AI in research are discussed throughout.

Topics Covered

• Foundations & Data: corpus creation, data exploration, keyness, and topic modeling (LDA/STM)
• Representing Meaning: lexicon-based methods, word embeddings, DDR, CCR, and BERTopic
• Applied Statistics: group comparisons, linear mixed models, and count models for text-derived measures
• LLMs & Misinformation: persuasiveness of LLM-generated content, linguistic complexity, and LLMs as simulated social agents
• Final Project: applied analysis on real-world data combining the methods covered throughout the course

Materials

Each session includes slides and Python or R scripts. Materials are added progressively as the course runs.