Recommender System Primer Flashcards Preview

11637 Foundations of Computational Data Science > Recommender System Primer > Flashcards

Flashcards in Recommender System Primer Deck (12)
Loading flashcards...
1

What data does collaborative filtering use?

Only uses Users' history of ordering or rating items

2

What are the data sources that can be used for a recommender system? (3)

  • Users' demographic information (e.g., age, gender, location)
  • Items' attributes (e.g., price, category)
  • Users' history of ordering or rating items

3

What do we know about data for collaborative filtering?

Very sparse

4

What are the representations of the collaborative filtering problem?

  • User-user approach: estimate a user’s rating of an item by finding "similar" users and then looking at their ratings for that item.
  • Item-item approach: estimate a user’s rating of an item by finding similar items and then looking at that user's rating of these similar items.
  • Matrix factorization: construct two low-rank matrices that approximate the observed entries of X.

5

What's The task of collaborative filtering?

"fill in" the missing values (i.e., the predicted user ratings) based on the existing ratings

6

How do you  convert a dataframe to a Numpy matrix?

df.to_numpy()

7

In the user-user approach, how is the similarity weight calculated?

Two common options:

  • Pearson correlation
  • Cosine similarity

8

  • What's a drawback of the user-user and item-item approaches?
  • How do you solve it?

  • Can't keep the user-rating matrix in sparse format, which does not scale well with a large number of users / items.
  • Use matrix factorization.

9

What's an advantage of the matrix factorization approach?

We don't need to break the sparsity of the user ratings matrix?

10

What is this? "singularity issues with matrix inverses"

11

  • What is alternating least squares?
  • What's an important point about it?

  •  
  • It's feasible with sparse matrices, which can considerably reduce runtime and memory usage.

12

What's matrix factorization?

This approach aims to approximate the observed entries in X as a product of two lower-rank matrices