Association analysis words

Association analysis words

1. AA transforms vast amount of data into a set of mathematically supported statements.

2. • Association rules

   • Are statements in the form
     • if antecedent(s) then consequent(s)
       • antecedent(s) -> consequent(s)
         • X -> Y
           • X,Y are itemsets
           • {T-shirt} -> {Lipstick}
             • {Skirt, Perfume} -> {Scarf, Nail Polish}
               • {Lipstick, Perfume -> {Skirt}
   • It is important to emphasise that association implies co-occurrence and not causality.

3. • Rule Support

   • Is the fraction of total transactions that contain both X and Y.
     • Support for {Skirt, Perfume, T-shirt} is 2/5.

4. • Confidence

   • Measures how often the items in the consequent set Y appear in antecedent transactions containing X
     • The association rule {Skirt, T-shirt} > {Lipstick},
       • X -> Y
     • Support count for (X,Y) = {Skirt, T-shirt, Lipstick} is 2
       • Because this occurs twice in the transaction sets (IDs 3 and 4)
     • Support count for X is 3
       • Because this itemset occurs thrice in the transaction sets (IDs 3, 4 and 5).

5. Itemset

   • Is a collection of one or more items;
     • {Skirt, Perfume, T-shirt} is an itemset.

6. • Support count

   • The frequency of occurrence of an itemset
     • Support count for {Skirt, Perfume, T-shirt} is 2
       • Because this itemset occurs twice in the transaction sets (IDs 4 and 5)