Courses
Courses at Sogang University in 2026, designed primarily for students in the Division of English and the humanities. Syllabi are available as PDF; schedules are tentative and subject to change.
Fall 2026
Linguistics with AI
This course explores the consilience of linguistics and artificial intelligence, examining how human language and its contexts can be understood and reimagined through computational tools. Rather than focusing on AI technology itself, it emphasizes how humanistic insight and creativity can guide the design of products and services people truly need.
- Format
- Lecture 70% · Discussion 10% · Presentations 20%
- Evaluation
- Midterm 30% · Final 30% · Presentations 10% · Projects 25% · Participation 5%
- Prerequisites
- None (basic NLP/AI concepts helpful)
Weekly topics (tentative)
- Introduction to AI and Linguistics — AI, NLP, GenAI
- Prompt Engineering — Generative AI (lab session, project team building)
- Syntax — Dependency Parsing
- Principles of Machine Learning (1) — Classification
- Principles of Machine Learning (2) — Word Embedding, RNN, LSTM, Transformer
- Principles of Machine Learning (3) — Transformer, Tokenization
- … semantics, pragmatics, human-centered AI applications, and team project presentations (see syllabus)
Data Analysis for AI
Designed for students in the English Division and the humanities, this course introduces fundamental experiences in data analysis. Students learn to handle textual data with Python, lower the barrier to coding through vibe coding, run preprocessing and exploratory analysis on large text datasets, and practice prompt engineering as a way of communicating with AI — integrating literary perspectives with data-driven methods.
- Format
- Lecture 45% · Experiment/Practicum 45% · Presentations 10%
- Note
- Discouraged for engineering students with coding experience. A paid AI tool (e.g. ChatGPT Pro, Claude Pro) is recommended for projects.
- Recommended
- Introduction to Linguistics (ENG2003), Linguistics with AI (ENG2112)
Weekly topics (tentative)
- Introduction to AI and Vibe coding 1 — Python basics (variables, strings, loops)
- Introduction to AI and Vibe coding 2 — Python basics
- Data Processing 1 — File I/O, string normalization, regex, tokenization
- Data Processing 2 — Sample datasets
- Data analysis with Python 1 — Pandas basics, word frequency, simple visualizations
- … exploratory text analysis, generative AI and prompt engineering, final projects (see syllabus)
Understanding and Using Generative AI
Designed for students without an AI background, this course introduces generative AI and other AI methods through intuitive examples. Students use tools such as ChatGPT, Gemini, and Claude, along with public AI-Hub datasets, to create value in their own fields through team projects—including business plans, research proposals, videos, and images. The emphasis is on understanding and application rather than implementing algorithms.
- Format
- Lecture 70% · Discussion 10% · Presentations 20%
- Evaluation
- Two midterm exams 60% · Discussion and presentations 10% · Team project 25% · Participation 5%
Weekly topics (tentative)
- Course orientation and an introduction to the history of AI (online lecture)
- Generative AI applications and prompt engineering (weeks 2–3)
- Machine learning fundamentals for developing generative AI
- Neural models for language: word embeddings and the evolution from RNNs to Transformers (weeks 6–7)
- Midterm exam 1
- Information retrieval and generative question answering
- Computer vision and multimodal AI
- Future directions: actionable AI and agents
- AI innovation in science, medicine, biology, law, manufacturing, and the arts · Midterm exam 2
- Critical evaluation and ethical use of AI outputs
- Final presentations of generative AI projects (weeks 14–16)
Spring 2026
Introduction to Data-Centric AI for Language Technology
Understand, evaluate, and improve Large Language Models from a "Data-Centric" perspective. Instead of model architecture or heavy coding, students learn to control AI with prompt engineering and to evaluate AI by constructing their own benchmark datasets, leveraging domain expertise in linguistics, culture, and education. The final goal is a KCI-level paper built on a domain-specific benchmark. Undergraduate and graduate sections are graded separately.
- Format
- Lecture 30% · Discussion 5% · Experiment 15% · Projects 35% · Presentations 15%
- Evaluation
- Midterm 15% · Quizzes 10% · Presentations 10% · Projects 40% · Assignments 20% · Attendance 5%
- Recommended
- Linguistics with AI (ENG2112), Data Analysis for AI (ENG3510), or equivalent
Weekly topics (tentative)
- Orientation and AI Literacy — Why Data Matters More Than Models?
- LLM Basics & Prompt Engineering — Instruction & Context
- Advanced Reasoning & Logic Design — Chain-of-Thought, Tree-of-Thought
- Evaluation and Benchmark — LLM Benchmarks, Linguistic Metrics, LLM-as-a-judge
- The Shadows of LLMs — Bias and Ethics, Hallucination
- … synthetic data generation, benchmark construction, and final research paper (see syllabus)