Summary on Recommendation Systems in Ecommerce Data Science

This brings us to the end of this session on recommendation systems. You started with understanding what recommendation systems are and why they are useful. Then, you moved on to the different types of recommendation systems: content-based filtering system and collaborative filtering system

In the content-based recommendation system, you will be recommended items that best match your taste. Thus, what matters in this context is your preferences and the qualities of the item in question. These are independent of the taste and preferences of any other user.

Collaborative filtering, on the other hand, takes into account the choices of other users. This again is of two types: user-based and item-based collaborative filtering. In user-based collaborative filtering, you will be recommended items that have been liked by other users who share your tastes and preferences. Item-based collaborative filtering, on the other hand, recommends items that are similar to the items you may have already rated or liked/disliked.