Numerical Problem


Question:

Consider an itemset I = X ∪Y = X′ ∪Y′ and let X′ ⊆ X. Show that confidence of the association rule

X ⇒ Y is greater than or equal to that of X′ ⇒ Y ′.

Answer:

Given:    I=X∪Y=X′∪Y′    and  X′⊆X

Proof:     Since,  I=X∪Y=X′∪Y′,  we know: Support(X∪Y) = Support(X′∪Y′)

This is because X∪Y and 
X' \cup Y'
represent the same set of items, so their support must be equal.

Given X′⊆XX' \subseteq X, it follows that:  Support(X′) ≥ Support(X).

This is because X′X' is a subset of XX, meaning every transaction that contains XX will also contain X′X'.

The number of transactions containing X' is at least equal to those containing X.

Comparing Confidence: ​

  1. Confidence of 𝑋⇒𝑌: Confidence(X⇒Y) = Support(X∪Y) / Support(X)
  2. Confidence of 𝑋′⇒𝑌′: Confidence(X′⇒Y′) = Support(X′∪Y′) / Support(X′)
Since, Support(X∪Y)=Support(X′∪Y′)\text{Support}(X \cup Y) = \text{Support}(X' \cup Y') and Support(X)≤Support(X′)\text{Support}(X) \leq \text{Support}(X') we have:

Confidence(X⇒Y)= Support(X∪Y)/Support(X) ​ ≥  Support(X∪Y)​/Support(X′) = Confidence(X′⇒Y′)

The confidence of the association rule X⇒Y is greater than or equal to that of X′⇒Y′X' \Rightarrow Y' when X′⊆XX' \subseteq X.
Raghunath

I am studying in M.SC Data Science at the Department of Computer Science and Engineering, Kalyani University. I am an enthusiast blogger.

Post a Comment

Previous Post Next Post