Collaborative filtering in ML Python - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is collaborative filtering in simple terms?

Collaborative filtering is a way to recommend things to you by looking at what other people with similar tastes liked.

beginner

What are the two main types of collaborative filtering?

The two main types are user-based filtering (finding similar users) and item-based filtering (finding similar items).

beginner

Why is collaborative filtering useful in real life?

It helps websites like Netflix or Amazon suggest movies or products you might like based on what others with similar tastes enjoyed.

intermediate

What is a common challenge with collaborative filtering?

A common challenge is the 'cold start' problem, where new users or items have little data, making recommendations hard.

intermediate

How does user-based collaborative filtering find recommendations?

It finds users who have similar preferences to you and recommends items those users liked but you haven't tried yet.

What does collaborative filtering mainly use to make recommendations?

A Data about user preferences and similarities

B Predefined rules about items

C Random item selection

D Only item features like color or size

Collaborative filtering uses data about what users like and finds similarities to recommend items.

Which type of collaborative filtering focuses on finding similar items?

A User-based filtering

B Item-based filtering

C Content-based filtering

D Hybrid filtering

Item-based filtering looks for items similar to those you liked before.

What problem occurs when a new user has no previous data in collaborative filtering?

A Overfitting

B Underfitting

C Cold start problem

D Data leakage

The cold start problem happens when there is not enough data about new users or items.

Collaborative filtering recommendations are based on:

A Random guesses

B Item features only

C User preferences only

D Both user preferences and item similarities

Collaborative filtering can use user preferences and item similarities to make recommendations.

Which of these is NOT a typical use case for collaborative filtering?

A Spam email detection

B Movie recommendations

C Product suggestions

D Music playlist creation

Spam email detection usually uses different methods, not collaborative filtering.

Explain how user-based collaborative filtering works in a recommendation system.

Think about how friends with similar tastes might suggest new things to you.

Describe the cold start problem in collaborative filtering and why it is challenging.

Imagine a new user who just joined and has no history yet.