Minimum Confidence and Minimum Apriori Support

Question

What are the corresponding values ​​for minimum confidence and minimum support values ​​for the Apriori algorithm ? How could you tune them? Are they fixed values ​​or do they change while the algorithm is running? If you used this algorithm before, what values ​​did you use?

Answer 1

I would suggest starting at 0.05 for support and 0.80 for confidence. But I agree that you must understand what exactly they represent in order to be able to correctly identify them. For rule A => B (where A, B are nonempty sets)

Support (A ⇒ B): s = P(A, B)
Confidence (A ⇒ B): c = P(B | A)
Lift (A ⇒ B): L = c/P(B)

Lifting is important for evaluating the rule of interest (because you usually come up with hundreds of them). More than twenty interesting considerations have been suggested. These include F-factor, kappa, mutual information, J-measure and Gini index. I personally order my rules according to the J-measure.

J.measure (A ⇒B): J = s/c * (c*log(L) + (1-c)*log((L-c)/L))

Answer 2

You must set the minsup and minconf values ​​before running the algorithm, and they do not change during the development process.

The choice of minsup options depends on your data.

80%. 0,05%. . , , , paterns.

, , . - 60%. .

, minsup, top-k mining. , , k = 1000, 1000 , , minsup. . TopKRules, . , , . : k minconf.