||MMDS: Beta Version of Third Edition
You are welcome to download components of what will become the third edition of
Mining of Massive Datasets by Jure Leskovec, Anand Rajaraman, and Jeff Ullman.
Please be advised that the new materials have not been reviewed, even by all of the
authors, and we would appreciate your telling us of any errors you find (email to
ullman at gmail dotcom).
Here is a table of the new materials and the major changes so far.
|1||Ch. 1||A revised discussion of the relationship between data mining,
machine learning, and statistics in Section 1.1.|
|2||Ch. 2||Spark and TensorFlow added to Section 2.4 on workflow
|3||Ch. 3||More efficient method for minhashing in Section 3.3|
|10||Ch. 10||More detail on finding overlapping communities in Section
10.5, approximate simrank and application to community-finding in Section 10.6,
and more efficient methods for parallel transitive closure in Section 10.8|
|12||Ch. 12||New section on decision trees|
|13||Ch. 13||New chapter on deep learning|
If you are an instructor interested in using the Gradiance Automated
Homework System with this book, start
by creating an account for yourself at
Then, email your chosen login and the request to become an instructor for the MMDS book
You will then be able to create a class using these materials.
Manuals explaining the use of the system are at
Students who want to use the Gradiance system for self-study can register at
Then, use the class token 1EDD8A1D to join the "omnibus class" for the MMDS book.
See The Student Guide for more information.