MultiwayClassification-package: Linear Classification of Multi-Way Data
In lockEF/MultiwayClassification: Linear Classification of Multi-Way Data

Description: Performs linear classification for data with multi-way structure. The distance-weighted discrimination (DWD) or support vector machine (SVM) classification objectives are optimized under the assumption that the multi-way coefficients have low rank.

Package:	MultiwayClassification-package
Type:	Package
Version:	1.0
Date:	2016-06-03
License:	GPL-3

Eric F. Lock and Tianmeng Lyu

Maintainer: Eric F. Lock <elock@umn.edu>

	data(IFNB_Data) ##Load gene expression time course data (?IFNB_Data for more info)
	
##multi-way DWD	
results.mw <- mul.dwd(DataArray,y=Class, rank=1) #estimate rank 1 model
names(results.mw)
##Compute projection onto the classification direction for each individual:
DWD_scores <- c()
for(i in 1:length(Class)) DWD_scores[i] = sum(DataArray[i,,]*results.mw$beta)+results.mw$int

##multi-way SVM
results.mw <- mul.svm(DataArray,y=Class,R=20, rank=1) #estimate rank 1 model
##Compute projection onto the classification direction for each individual:
SVM_scores <- c()
for(i in 1:length(Class)) SVM_scores[i] = sum(DataArray[i,,]*results.mw$beta)+results.mw$int