Recommendation Systems

1. Collaborative Filtering:

Collaborative Filtering recommend items based on crowdsourced information about user’s preferences for items

* user based 
* items based

1.1 Item based

below is an example of item based recommender:

userD and userB gave 4 stars to the cellphone and the cellphone case, and userA gave 4 stars to the cellphone. So the model would recommend the cellphone case to userA.

1.2 User based

userB is similar to userD, userD likes his life insurance, so let’s recommend it to userB also.

2. Cotent-based Recommender

Content-based recommender recommend items based on similarities between features.

3. Popularity based recommenders

Based on simple count statistics but don’t take user