Computer Science: Data Mining Help

[miller.4314]

I need help with a problem in my CS420 Data Mining course. The question asks to use the Relief Algorithm for feature selection. The algorithm is as follows:

 

W(A­_i)_new = W(A_i)_old – (diff(X[A_i], H[A_i])² + diff(X[A_i], M[A_i])²)/m

 

where W(Ai) is the quality score for all feature Ai, X is a randomly selected sample, H is nearest Hit, M is nearest Miss, and m is the number of randomly selected training samples.

 

Any help with this problem would be greatly appreciated. Thanks!