What is support and confidence in data mining?

Question

Can anyone explain the support and confidence in data mining?

Answer 1

The support and confidence terms are used in implementing Market basket analysis. These helps in identifying the joint purchasing and associations between products.

Support represents the popularity of that product of all the product transactions. Support of the product is calculated as the ratio of the number of transactions includes that product and the total number of transactions.

Support of the product = (Number of transactions includes that product)/ (Total number of transactions)

Confidence can be interpreted as the likelihood of purchasing both the products A and B. Confidence is calculated as the number of transactions that include both A and B divided by the number of transactions includes only product A.

Confidence (A=>B) = (Number of transactions includes both A and B)/ (Number of transactions includes only product A)

If you wish to learn Data mining and predictive analytics using R, I suggest these Data Scientist by Intellipaat.

Also, watch this video on Data Mining:

Answer 2

Support - It refers to the number of transactions that contain a particular item. 

support(A) = Number of transactions containing A / Total number of transactions.

Confidence - If you have to find the confidence for a rule A-> B, it means you are required to know how often B appears in a transaction that contains A. 

Confidence (A->B) = 

Number of transactions containing A and B / Number of transactions containing A.