Research Ideas

Bayes Nets

Nir Friedman and Moises Goldszmidt have produced this excellent (and long) tutorial on Learning Bayesian Networks from Data

Hubs and Authorities

Jon Kleinberg gave an excellent theory seminar here on the notion of hubs and authorities on the World Wide Web. I think the paper on the topic is readable enough - even though I can't grok eigenvectors and matrices all that well.

PageRank

A gaggle of Google founders came up with another very cute iterative algorithm to apply to Web pages. PageRank assigns an importance score to each Web page based on the stationary distribution of a random walk with random restarts on the Web.

Stochastic Hubs and Authorities

Ronny Lempel and Sholmo Moran took the random walk idea present in PageRank and applied it to Hubs and Authorities. They called it SALSA and explained how it's better than Hubs for some graphs. A second version was submitted to the ACM Transactions on Information Systems