Get started with user click analysis for ecommerce recommendations

Question

I am building a website similar to an eCommerce site. I want to implement personalization to that. To achieve this, I've been told that click stream analysis will be better. As data I plan to collect logged in user ID, click time of the day, price of the clicked item. Otherthan that user's personal information such as gender, location will be used. I've read many articles and tutorials about recommendation systems. But they are based on user ratings of an item. In my case, users cannot rate items. And so I find it difficult to get started. I would like to know about how to get started with this. How can I use this collected data to come up with user recommendations for a particular user??

Any help would be much appreciated.

Answer 1

you can use binary rating from users.each user that click each item you have a rate 1 and each item that you recommend user and he refuse it you earned 0 rate. for more information and make online recommendation look at this paper.