ECLAT

An unsupervised machine learning model used for Association Rule Learning and Frequent Itemset Mining that uses a vertical data format and Depth-First Search strategy.

a) ECLAT Algorithm b) Apriori Algorithm c) FP-Growth Algorithm d) K-Means Clustering

a

The data format where each row represents a unique Transaction ID (TID) and columns contain the items purchased.

a) Vertical Data Format b) Horizontal Data Format c) Matrix Data Format d) Longitudinal Data Format

b

The data format where each row represents a unique item and the column contains a list of all TIDs in which that item appears.

a) Horizontal Data Format b) Matrix Data Format c) Vertical Data Format d) Longitudinal Data Format

c

The list of all Transaction IDs in which a specific item appears is called a:

a) Itemset list b) Support vector c) TID-set d) Frequency array

c

The search strategy used by ECLAT to find frequent combinations once the dataset is in vertical format.

a) Breadth-First Search (BFS) b) Best-First Search c) Depth-First Search (DFS) d) A* Search

c

The search strategy used by the traditional Apriori algorithm, which ECLAT differs from.

a) Depth-First Search (DFS) b) Uniform Cost Search c) Breadth-First Search (BFS) d) Greedy Search

c

The primary metric used in ECLAT to filter out infrequent combinations, measuring the absolute frequency or proportion of transactions containing a specific itemset.

a) Confidence b) Lift c) Support d) Conviction

c

The formula for Support is Frequency(X) divided by:

a) Total items b) N (total number of transactions) c) Number of frequent itemsets d) Minimum support threshold

b

The core mathematical operation in ECLAT, used to find the transactions containing multiple items simultaneously.

a) Set union b) Set difference c) Cartesian product d) Set intersection

d

The TID-set intersection for items A and B is mathematically expressed as:

a) TID(A) ∪ TID(B) b) TID(A) ∩ TID(B) c) TID(A) × TID(B) d) TID(A) − TID(B)

b

The frequency of the itemset {A ∪ B} is calculated as:
a) |TID(A) ∪ TID(B)|
b) |TID(A) ∩ TID(B)|
c) |TID(A)| + |TID(B)|
d) |TID(A)| × |TID(B)|

b

In the horizontal example provided, the items in Transaction T1 are:

a) {Apple, Bread} b) {Bread, Milk} c) {Apple, Bread, Milk} d) {Apple, Milk}

c

In the vertical example provided, the TID-set for Apple is:

a) {T1, T2, T3} b) {T1, T3} c) {T1, T2} d) {T2, T3}

c

In the vertical example provided, the TID-set for Bread is:

a) {T1, T2} b) {T2, T3} c) {T1, T2, T3} d) {T1, T3}

c

In the vertical example, the intersection of TID(Apple) and TID(Bread) is:

a) {T1, T3} b) {T1, T2} c) {T2, T3} d) {T1, T2, T3}

b

The size of the intersected set for {Apple, Bread} represents how many times the itemset appears, which is:

a) 1 b) 3 c) 2 d) 4

c

The event or condition that logically precedes another in an "if-then" rule, representing the initial item a customer puts in their basket.

a) Consequent b) Support c) Antecedent d) Confidence

c

The outcome that follows as a direct result of a prior action, representing the "then" portion of an association rule.

a) Antecedent b) Support c) Consequent d) Lift

c

The metric that represents the conditional probability of a rule, measuring how often the consequent is purchased given the antecedent is already in the transaction.

a) Support b) Lift c) Frequency d) Confidence

d

The metric that measures the predictive strength of an association rule over random chance, calculating how much more likely a customer is to buy the consequent when they buy the antecedent.

a) Support b) Confidence c) Lift d) Frequency

c

According to the lesson objectives, students should be able to differentiate between horizontal and vertical data formats in:

a) File systems b) Programming languages c) Database structures d) Operating systems

c

The ECLAT algorithm is mathematically grounded in set theory and the calculation of:

a) Confidence b) Lift c) Support d) Entropy

c

ECLAT stands for Equivalence Class Transformation.

True

ECLAT uses a horizontal data format and a Breadth-First Search strategy, similar to the Apriori algorithm.

False

In the vertical data format, each row represents a unique item and the column contains a list of all the TIDs in which that item appears.

True

ECLAT uses set union to find the transactions containing multiple items simultaneously.

False

The Support formula is Support(X) = Frequency(X) / N, where N is the total number of items in the database.

False

ECLAT scans the original horizontal database multiple times during the mining process.

False

One advantage of ECLAT is that it avoids the massive combinatorial explosion of candidate sets that occurs in Breadth-First Search algorithms.

True

ECLAT natively calculates Confidence and Lift to generate "If-Then" association rules.

False

A limitation of ECLAT is that intersecting massive TID-sets in dense datasets can consume a large amount of RAM.

True

An antecedent is the "then" portion of an association rule.

False

Lift measures the conditional probability of a rule.

False

One application of ECLAT listed is Web Usage Mining, where web administrators analyze clickstream logs.

True

In the set intersection example, the intersection of TID(Apple) = {T1, T2} and TID(Bread) = {T1, T2, T3} yields {T1, T2, T3}.

False

The ECLAT algorithm continues building larger itemsets recursively until the frequency falls below a predefined minimum threshold.

True

One advantage of ECLAT is that it requires only a single database scan.

True

A limitation of ECLAT is the lack of native rule generation.

True

In Market Basket Analysis, ECLAT helps retailers discover products frequently bought separately.

False

In Medical Diagnosis Clustering, patients are treated as transactions and symptoms as items.

True