UCSD's Analyzing Unstructured Data Course: From Text Mining to LLMs
caglar_ee · x · 2026-08-06
Video lectures for UC San Diego's graduate course MGTA 415: Analyzing Unstructured Data (Winter 2025), taught by Prof. Jingbo Shang, are now available online.
The curriculum covers a comprehensive mix of traditional and modern techniques:
- Basics: Linear/logistic regression and text preprocessing (tokenization, POS tagging, parsing).
- Core NLP: Text classification, information retrieval, topic modeling, and word embeddings.
- Advanced Mining: Phrase mining, NER, and taxonomy construction.
- Weak Supervision: Emphasizes unsupervised, weakly, and distantly supervised methods like bootstrapping and learning from seed words.
- LLMs: Offers a formula-free explanation of ChatGPT and practical insights on utilizing large language models.
Related event: UCSD Releases Advanced Text Mining Course Videos(2 posts)→
More from Research
- Paper Proposes Token-Native Storage Architecture for AI Agents — bclavie · 2026-08-06
- SJTU's ABSeeker: 4B Search Agent Matches 30B Models via Step-Level Credit Assignment — SJTU · 2026-08-06
- FocusMem: Factorizing Latent Memory in GUI Agents with a Trust Gate — Zhuoran Zhang · 2026-08-06
- Survey on Self-Evolving Coding Agents: Taxonomy and Challenges — NJUST1412 · 2026-08-06
- SREGym: Evaluating SRE Agents with High-Fidelity Production Faults — tianyin_xu · 2026-08-06
- Horizon Robotics Launches EmbodiedGen V2: Natural Language to Simulation-Ready 3D Worlds — rsasaki0109 · 2026-08-06