问题
When the percentage values of support and confidence is given how can I find the minimum support in Apriori algorithm. For an example when support and confidence is given as 60% and 60% respectively what is the minimum support?
回答1:
The support and confidence are measures to measure how interesting a rule is.
The minimum support and minimum confidence are set by the users, and are parameters of the Apriori algorithm for association rule generation. These parameters are used to exclude rules in the result that have a support or a confidence lower than the minimum support and minimum confidence respectively.
So to answer your question, when you say that: "For an example when support and confidence is given as 60% and 60% respectively what is the minimum support?" you probably mean that you have set the minimum support and confidence to 60 %.
I think that you are just confused by the terms.
回答2:
Check out the full explanation of Apriori algorithm with live usable example here:
http://www.codeding.com/articles/apriori-algorithm
You can add new items and enter the minimum support threshold and minimum confidence threshold and see the resultant large itemsets generated instantly in the demo Silverlight widget.
回答3:
My answer is coming a little too late, but I guess what Chanikag is asking is - "How to minimum support count when the support threshold is given as 60%". The Minimum Support Count would be count of transactions, so it would be 60% of the total number of transactions. If the number of transactions is 5, your minimum support count would be 5*60/100 = 3.
回答4:
Minimum support count is the % of the all transaction.suppose you have 60% support count and 5 is the total transaction then in number the min_support will be 5*60/100=3.
回答5:
I'm not sure your question makes sense. From your example if you have at least one rule returned with Support and Confidence of 60% you can be sure that the Minimum-Support is at least 60% but it could be more.
Minimum-Support is a parameter supplied to the Apriori algorithm in order to prune candidate rules by specifying a minimum lower bound for the Support measure of resulting association rules. There is a corresponding Minimum-Confidence pruning parameter as well.
Each rule produced by the algorithm has it's own Support and Confidence measures. Roughly, Support is the ratio of instances for which the rule is true out of all instances. Confidence is the ratio of instances for which the rule is true out of the number of instances for which the antecedent (LHS of the implication) is true.
Check out Wikipedia for more rigorous definitions.
来源：https://stackoverflow.com/questions/10364632/how-to-find-the-minimum-support-in-apriori-algorithm