Statistical Foundations of Natural Language Processing

EEE 486 treats language as data you can model probabilistically — the course is built around the idea that meaning, structure, and ambiguity in text can be captured with statistics, vector representations, and sequence models like HMMs rather than hand-written grammar rules. You will work through three projects plus a term project that take you from tokenization and n-gram models up to word embeddings and neural language models, applying the math (probability, hypothesis testing, estimators) directly to real corpora. It sits at the intersection of EEE's signals/statistics background and modern ML, giving you the foundations needed before jumping into transformer-era NLP or computational social science research.