Hi Rodrigo, There's a complete field: "Data Mining" that deals with questions like these. There are a class of algorithms called "recommender" algorithms that analyze past purchase history of a user to make future predictions of items of interest (similar to the ones implemented on Ebay, Amazon etc. where they earn big bucks by figuring out what a customer would buy). However, if you're inexperienced or looking at just starting out with simple recommendations, I would suggest that you sit down with ecommerce data from your site and make the following analysis manually:
1. What items are frequently bought together? Place them in close proximity on your site. (See the Apriori and FP Tree algorithms for simple algorithms for the same. You could also look at clustering algorithms if you have data that is amenable to such analysis) 2. If a user has bought certain items, what are the items he would be likely to be interested in? (See Classifiers of various kinds to derive predictive rules from the frequently bought together dataset of step 1 - Bayesian Classification, Rule Based Classification, Decision Tree classification - there are a number of options available). I hope this helps. Let me know if you need more information about this. -- DK Follow me on Twitter: http://twitter.com/divyekapoor http://www.divye.in -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To view this discussion on the web visit https://groups.google.com/d/msg/algogeeks/-/XMGrNGn55k4J. To post to this group, send email to algogeeks@googlegroups.com. To unsubscribe from this group, send email to algogeeks+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.
