Lecture, three hours. Requisites: courses 210A, 210B, or consent of instructor. Conceptual, mathematical, and computational foundations of machine learning, with special focus on social science applications. Survey of supervised and unsupervised methods, including Naïve Bayes, k-means, logistic regression, decision trees (classification and regression), topic models, and neural networks. Practicalities of implementation on range of data types. S/U or letter grading.