|
Language of Instruction
|
English
|
|
Level of Course Unit
|
Bachelor's Degree
|
|
Department / Program
|
Political Science and Public Administration (English)
|
|
Type of Program
|
Formal Education
|
|
Type of Course Unit
|
Elective
|
|
Course Delivery Method
|
Face To Face
|
|
Objectives of the Course
|
By the end of the course, students will be able to:
1. Explain core capabilities/limitations of modern LLMs in a social-science context.
2. Design structured prompts and multi-step task plans to obtain reliable outputs.
3. Mitigate hallucinations through sourcing, triangulation, and verification workflows.
4. Select appropriate models by task (accuracy vs. speed, context window, modality, privacy needs).
5. Use Deep Research features to plan and execute traceable desk research.
6. Work in a Canvas environment to build and manage multi-artifact projects (notes, drafts, slides).
7. Create and govern custom GPTs and evaluate existing ones for safety, privacy, and utility.
8. Deliver a complete social-science mini-project that is transparent, replicable, and ethically sound.
|
|
Course Content
|
This practice-oriented course introduces how large language models (LLMs) and AI assistants can be used rigorously and responsibly in social-science work. We focus on three model families—ChatGPT, Claude, and Gemini—and cover prompt design, task decomposition, model selection by task, hallucination-reduction and verification methods, Deep Research workflows, using a Canvas-style environment for multi-file projects, creating custom GPTs, and evaluating/using existing custom assistants. Applications include literature reviews, research design, qualitative coding, survey drafting, memo/policy writing, and project scaffolding.
|
|
Course Methods and Techniques
|
|
|
Prerequisites and co-requisities
|
None
|
|
Course Coordinator
|
None
|
|
Name of Lecturers
|
Associate Prof.Dr. BAŞAR BAYSAL
|
|
Assistants
|
None
|
|
Work Placement(s)
|
No
|
Recommended or Required Reading
|