In brief:
I have put, on pages in a store, a javscript that creates a unique ID for any user, browsing the site, then, stores the product refs of each view. Every time a user views a product, a row is added. if they have looked at it before, the ViewNum is incremented for that UID/ref.
Here is my table:
So, what i'm looking for, is to use the data set, to work out what products are related, based on those products being viewed also by other people, and what they looked at.
I have looked at neural networks, and some rather crazy sql.
can anyone shed some light on this one? I'v struggled with it before and never got anywhere.
I have put, on pages in a store, a javscript that creates a unique ID for any user, browsing the site, then, stores the product refs of each view. Every time a user views a product, a row is added. if they have looked at it before, the ViewNum is incremented for that UID/ref.
Here is my table:
Code:
CREATE TABLE IF NOT EXISTS `actTrak` ( `ID` int NOT NULL AUTO_INCREMENT, `UID` text, `ProdID` text, `ViewNum` int, PRIMARY KEY (`ID`) ) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;
I have looked at neural networks, and some rather crazy sql.
can anyone shed some light on this one? I'v struggled with it before and never got anywhere.
Comment