KOS_Data: A list consisting of Training and Test data along with...

In biClassify: Binary Classification Using Extensions of Discriminant Analysis

Description

A list consisting of Training and Test data along with corresponding class labels.

Usage

KOS_Data

Format

A list consisting of:

TrainData

(179 x 4) Matrix of training data features. the first two features satisfy sqrt(x_i1^2 + x_i2^2) > 2/3 if the ith sample is in class 1. Otherwise, they satisfy sqrt(x_i1^2 + x_i2^2) < 2/3 - 1/10 if the ith sample is in class 2. The third and fourth features are generated as independent N(0, 1/2) noise.

TestData

(94 x 4) Matrix of test data features. the first two features satisfy sqrt(x_i1^2 + x_i2^2) > 2/3 if the ith sample is in class 1. Otherwise, they satisfy sqrt(x_i1^2 + x_i2^2) < 2/3 - 1/10 if the ith sample is in class 2. The third and fourth features are generated as independent N(0, 1/2) noise.

CatTrain

(179 x 1) Vector of class labels for the training data.

CatTest

(94 x 1) Vector of class labels for the test data.

...

Source

Simulation model 1 from [Lapanowski and Gaynanova, preprint].

References

Lapanowski, Alexander F., and Gaynanova, Irina. “Sparse Feature Selection in Kernel Discriminant Analysis via Optimal Scoring”, preprint.

biClassify documentation built on Dec. 11, 2021, 9:22 a.m.