do anyone know theory of laplace,gain,p-s,conviction in association rule

hi all
does anyone know tutorial how to count laplace,gain,p-s,conviction in association rule. As I know association rule, usually use support,confidence and lift .But when I generate rule from rapidminer ,i got 4 more variables and i don't understand what it used for and how to count.
tHANK You for your attention and I hope someone can help me with this one.

Find more posts tagged with

AI Studio

Association Rules + Mining

Accepted answers

All comments

deniz

I need help for laplace, gain, p-s, conviction and lift, please help me. I need brief information.

deniz

dear all
please help me I really need to what they mean, I made an analysis but I cannot comment them.

edegraaf

gain is probably the information gain as used in decision trees: entropy before spliting on the rule - entropy after spliting on the rule (http://en.wikipedia.org/wiki/Information_gain_in_decision_trees). Conviction is well described under Association Rule mining when you use wikipedia. Now laplace, I am not sure, the definition of laplace:
Laplace Criterion: If the probabilities of several chance events are unknown, they should be assumed
equal, and the different actions should be judged according to their payoffs averaged over all the states
of nature. The question is what is the payoff?