Feature-Weighted Naive Bayesian Classifier for Wireless Network Intrusion Detection

Algorithm 1

The JINB algorithm.

	Input: training sample set U, sample instances to be classified , and category label
	Output: sample J belongs to category
	z = number of categories
	Obtain the prior probability U (cx)
	for each x in z
	t = 0
	s = 0
	for each x in U
	t = t + 1
	if (I ∈ cx) s = s + 1
	end for
	U (cx) = s/t
	for each x in z
	U (J | cx) = 1
	for each y in
	weight (x, y) = WJS (x, y)  WICF ()
	U (J | cx) = U (J | cx)  U ( | cx)  weight (x, y)
	end for
	ux = U(cx)  U (J | cx)
	end for
	end for
	output ()