confNetFunc {BiCausality}

confNetFunc function

Description

This function Computes a confidence network in data mining. Given a set of n transactions or samples in mat s.t. each transaction has d binary items. The conf(mat[,j]=1|mat[,i]=1) is a ratio of a number of samples in jth and ith dimensions that have values equal to one divided by a number of samples in the ith dimension that has a value equal to one. The confNetFunc computes the network where the nodes are dimensions and the edge weights are conf(mat[,j]=1|mat[,i]=1) for any directed edge from i to j.

Usage

confNetFunc(mat, ths = 0.1)

Arguments

mat

is a matrix n by d where n is a number of transactions or samples and d is a number of dimensions.

ths

is a threshold parameter for cutting of the edge weights. There exists the directed edge from i to j if its edge weight if above or equal ths.

Value

This function returns a binary adjacency matrix confNet and the weighted adjacency matrix confValMat.

confNet

A binary adjacency matrix that has confNet[i,j]=1 if confValMat[i,j]>=ths. Otherwise, it is zero.

confValMat

A weighted adjacency matrix where confValMat[i,j] is conf(mat[,j]=1|mat[,i]=1).

Examples

res<-confNetFunc(mat)