.. so the equivalent there would be to pre-compute all item-item similarities using, say, LogLikelihoodSimilarity, then load those into memory by stuffing them into a GenericItemSimilarity. Then plug that in to a GenericItemBasedRecommender along with a MySQLBooleanPrefJDBCDataModel pointing to a database with all your data, and should work out reasonably.
You can pseudo-distribute this arrangement too. These are your non-distributed options -- I'd still say for this size, the proper distributed version is perhaps better to start looking at. But options are good. On Wed, May 5, 2010 at 11:59 AM, First Qaxy <qa...@yahoo.ca> wrote: > Thanks! > Slope-one is on the map when I'll start to look into recommending based on > user satisfaction(ratings). At this point I'm focusing on user interest which > limits me to boolean based algorithms. > -qf
