6: Market Basket Analysis and Text Analysis Flashcards

1
Q

What does a leverage value of 0 indicate?

A

A leverage value of 0 indicates independence between A and B.

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

Between which values does the lift measure range?

A

From 0 to infinity.

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

Between which values does leverage measure range?

A

Between -1 and 1.

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

What does it mean if the measure lift(A→B) = 3.5?

A

That a customer who buys item A will be 3.5 times more likely to buy B (at same purchase).

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

What does it mean if the measure Lift(A→B) > 1?

A

That the presence of A in this transaction causes a higher probability of B in the same transaction.

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

What does the lift (A→B) measure compare?

A

It compares P(B∣A) with P(B), i.e, the probability of B occurrence given A is present and the probability of B occurrence without knowing about A.

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

Definition of support A & B?

A

The proportion of times events A and B occurred together is referred to as the support of A and B.

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

How do we measure co-incidences?

A

We look for products who have 1s at the same row (transaction) in the incidence matrix.

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

Definition of confidence?

A

Confidence measures the percentage of times that item B is purchased, given that item A was purchased. (Confidence of an association rule is a measurement of its predictive power or accuracy).

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

Formula for confidence of B given A?

A

Confidence(A→B) = P(B|A) = support (A & B) / support (A) = (count (A & B) / N) / (count (A) / N)

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

What is the idea of using the lift measure and not only the confidence measure (A→B)?

A

If B (following itemset = consequent) is a very frequent itemset, confidence related to B will always be high.) Lift is another metric to be considered together with confidence.

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

Formula for lift of A on B?

A

Lift(A→B) = P(B|A) / P(B)

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

What does the leverage of A and B measure?

A

The strength of association between A and B.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What does the conditional probability of B given A express?

A

the probability of event B, knowing that event A has occurred.

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

Which two parts does an association rule consist of?

A

The antecedent → the consequent.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Give some examples of how MBA can lead to better decision making within a company?

A

1) By being able to make intelligent assortment decisions based on what items sell together, 2) by giving a complete analysis of promotion effectiveness, 3) Better item pairings in marketing materials and store displays, 4) Improved inventory distribution by store

17
Q

What does it mean to have high co-incidences?

A

A large relative frequency of 1s in the incidence matrix.

18
Q

What does it mean for leverage to be greater than 0?

A

That A has a lift on B, and B has a lift on A.

19
Q

Formula for leverage of A on B?

A

P(A snitt B) - [P(A) P(B)] = support (A,B) - [(support(A) * support(B)]