Data sets for the Book

Modeling the Internet and the Web

Probabilistic Methods and Algorithms

Pierre Baldi, Paolo Frasconi, Padhraic Smyth

We very much appreciate the suggestion of pointers to other related data sets and research software. Please click one of the authors' names above to reach his webpage and send comments.

• Prescript (PostScript to text) utility, maintained at the New Zealand Digital Library website.
• The stemming algorithm of M. Porter, maintained by the author.
• The TREC test collection, maintained by NIST.
• The Reuters-21578 data set, maintained by D. Lewis.
• The 20 newsgroups data set , maintained by J. Rennie.
• The SVM-light software, maintained by T. Joachims.

• The Server-side session-length data for Figures 7.6 through 7.9
• Sequences of page requests for msnbc.com used for fitting mixtures of Markov models, described in Section 7.4.4