Skip to main content

Table 2 Notations

From: Min-redundancy and max-relevance multi-view feature selection for predicting ovarian cancer survival using multi-omics data

Symbol Definition and Description
D =  < X, y> Labeled dataset where XRm × n is a matrix of m instances and n features, and y {0, 1}m is the binary class labels of the instances
xi ith feature in X
g(xi, xj) Function that returns the redundancy between two features xi and xj
f(xi, y) Function that returns the relevance between a feature xi and class labels y
S Indices of selected features
Ω Indices of all features
ΩS Indices of candidate features Ω − S
k Number of features to be selected
v Number of views in a multi-view dataset
MVD =  < (X1, …, Xv), y> Labeled multi-view dataset where \( {X}^i\in {R}^{m\times {n}_i} \) is a matrix of m samples and ni features and y {0, 1}m is the binary class labels of the instances in all views
Di =  < Xi, y> ith view in a multi-view dataset
\( {\mathrm{x}}_{\mathrm{j}}^{\mathrm{i}} \) jth feature in Xi
Si Indices of selected features from ith view
Ωi Indices of all features in ith view
\( {\Omega}_{S^i} \) Indices of candidate features Ωi − Si in ith view