Database Matching Algorithm

I am working on a system which will match members based on exact or/and closeness of their choice of product catego开发者_开发技巧ries.

For eg. Would suggest to a member other members whom have the same choice or likely or close selections.

Eg. Fred selects (Cereals, Fruits, Beverages, Apparel) Mike selects (Cereals, Fruits) Jane selects (Fruits, Beverages, Apparel)

I want to match and suggest to Fred that a member Jane has categories in common

Any pointers?

I would suggest this book: http://www.amazon.com/Programming-Collective-Intelligence-Building-Applications/dp/0596529325. I have it, and it has many examples of useful algorithms that do exactly what you want. In fact, the second chapter is all about making recommendations.

Although the primary language it uses is Python, it isn't too difficult to apply the examples to PHP.

Do a query on the user's categories where category = '$category_id1' and category = '$category_id2', etc. You will also need to put the categories into sub categories so that if they belong to the same parent category they are considered similar. (i.e. Tops -> Tees, Graphic Tees, Long Sleeve Shirts, Sweaters...)