DHUM 25A43 - Investigating with AI

How can we leverage cutting-edge AI tools to produce digital investigations for social sciences ? In this course, students will master the use of Large Language Models (LLMs) and AI agents for data collection, analysis, and visualization. The curriculum goes beyond traditional data analysis by teaching students how to extract hidden patterns and uncover semantic meanings from dynamically built corpora. This approach enables them to gather insights and design results that can only be achieved with the assistance of AI tools. Students will gain an understanding of both the capabilities and limitations of AI tools in research contexts. They will explore how AI tools can enhance our understanding of social phenomena and examine the strengths and limitations of AI-assisted research. Through hands-on projects, students will investigate these questions using real-world data from various sources such as climate negotiations reports, parliamentary speeches, social media discourses etc. Additionally, students will learn to develop their own methodological toolbox by generating an analysis pipeline with the assistance of AI. The course goes over fundamental concepts in Machine Learning and Natural Language Processing and culminates in a student-led investigation project. Working in groups, students develop their own research protocol, collect and analyze data using AI tools, and present their findings in the form of a website. Each student will also produce an individual reflective essay on his/her experience with AI-assisted research.

Alexis PERRIER,Andrei MOGOUTOV

Séminaire

English

Beginner level in python is recommended but not mandatory in order to have a minimal understanding of the code produced by AI agents; experience using chatGPT, claude.ai or similar public chatbots

Spring 2024-2025

Students first develop a research protocol to collect and analyze online data about a public issue of their choice, presenting their findings through a website. Each step of this investigation must demonstrate the appropriate use of AI tools. Additionally, students submit an individual reflective note examining their experience with AI-assisted research, specifically analyzing how their practical use of AI tools aligned with the course's theoretical framework.