Ted coming back to your message here as well -- let me see if I understand how to apply this.
Imagine we know which users have seen which movies (no ratings, just the set of movies). For each movie I know how many users have seen both, or only one, or neither. I can use this to figure out pairs of movies that are seen together unusually often. Then for each user, find the movies among those they haven't seen which are most likely to occur together with the ones they have seen, and recommend them. On Fri, May 30, 2008 at 9:13 PM, Ted Dunning <[EMAIL PROTECTED]> wrote: > I am a strong proponent (not surprisingly) of using log likelihood ratio > (LLR) tests as a primary filter for finding good matches in cooccurrence > analyses like this. It is very simple and has proven effective for many > years in work I have done as well in other peoples work. Correlation based > measures are very bad at extreme counts. > > See here for a bit more detail: http://citeseer.ist.psu.edu/29096.html . I > should have a blog entry up which illustrates how simple LLR tests are to > implement (literally just a few lines of R and only twice or four times that > many in Java). > > LLR can also be used very effectively for document routing using what are > essentially variants on Naive Bayes classifiers where features are selected > using LLR and weighted using corpus frequencies (usually something like log > IDF). Other classifiers where you have vast numbers of possible features > should work well based on the same work.
