Lecture 5 Flashcards

1
Q

recommendation systems

A

help consumers find products that match well with their preferences

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How are long tail and recommended systems related

A

consumers may be aware of some products but typically those are the ones available in the head of the tail

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What type of data is in RS

A

Explicit data (ratings)
Implicit data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Explicit data

A

people are asked explicitly to rate items

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

types of recommendation systems

A

1.content-based
2.collaborative-filtering

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

content based-recommendation systems

A

making recommendations based on past actions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

collaborative-filtering systems

A

based on activities and preferences of other users similar to you

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Data in RS

A

Explicit and Implicit

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

how to build a profile of a user

A

E (attribute score * normalized rating )/ sum of attribute score

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How to build recommendation system

A

by Cosine Similarity (the dot product for normalized vectors)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How does RS work?

A

work based on the similarity between either the content or the users who access the content

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Interpreting results from Cosine Similarity

A

0 means no similarity, whereas 1 means a match

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Problems with RS

A
  • causality, because indeed prediction is being measured
  • consider the counterfactual effect: what would have happen if the user not seen the recommendation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

When the Recommendation system is accurate but useless

A

doing the groceries
RS will recommend bananas, but without the system I would buy them anyway.
The RS is 90% accurate but the incremental value is 0, will not change my behavior

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Calculating content-based filtering

A
  1. Normalize rating
  2. Build a user profile
  3. Cosine similarity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Calculate collaborative filtering

A
  1. calculate correlation between users
  2. approximate the numbers