There's tons of material online but yet I can't reconcile the different definitions for recommender system methodologies / strategies.
I think we can identify several axes:
- memory vs model based; Model based models do not need to store historical data such as user ratings or proxies (implicit data) such as purchase behavior, views, etc.
- product based vs collaborative filtering, aka user-based filtering. Collaborative filtering relies on user preferences data (implicit or explicit - explicit when users explicitly provide ratings, likes, stars). Product based rely uniquely on product features.
collaborative filtering comprises item based, which is totally based on user-item rankings, and content-based recommendation, which exploits product features and user preferences (although other sources put content based filtering outside the family of collaborative filtering)
Is this correct? Anything (big) missing?
