Mining Massive Datasets

Mining Massive Datasets is the StanfordOnline edX course taught by Jure Leskovec, Anand Rajaraman and Jeff Ullman, based on their textbook of the same name, which the page says can be downloaded free by arrangement with the publisher; the content closely matches Stanford's CS246. The major topics are MapReduce systems and algorithms, locality-sensitive hashing, algorithms for data streams, PageRank and web-link analysis, frequent-itemset analysis, clustering, computational advertising, recommendation systems, social-network graphs, dimensionality reduction and machine-learning algorithms. The page states that the course is intended for graduate students and advanced undergraduates in computer science and expects prior courses in data structures, algorithms, database systems, linear algebra, multivariable calculus and statistics; its FAQ says workload varies but ten hours a week is a good guess.

edX lists it as self-paced, advanced level and seven weeks at five to ten hours a week. Auditing is free with time-limited access and no graded work or certificate; the verified track with an edX certificate costs $149.