In seldas/Dforest: Decision Forest

Dforest

Introduction

Dforest is a R-implementation of Decision Forest algorithm that combines the predictions of multiple independent decision tree models for a consensus decision. In particular, Decision Forest is a novel pattern-recognition method which can be used to analyze:

1. DNA microarray data;
2. Surface-Enhanced Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (SELDI-TOF-MS) data; and
3. Structure-Activity Relation (SAR) data.

In this package, three fundamental functions are provided, as

1. DF_train,
2. DF_pred, and
3. DF_CV.

run Dforest() to see more instructions.

Reference
Tong, Weida, et al. "Decision forest: combining the predictions of multiple independent decision tree models." Journal of Chemical Information and Computer Sciences 43.2 (2003): 525-531.

Index of Functionality

First, simply install and load Dforest package into your R environment:

if (!require(Dforest)){
  install.packages("Dforest",repos = "https://cran.r-project.org/")
}
library(Dforest)

Training Decision Forest Model - DF_train

DF_train() is used to train Decision Forest model based on training dataset.
Here is an simple example for using this function.

  data(demo_simple)
  set.seed(188)
  # for simplicity, we only used two-classes classification.
   X <- subset(data_dili$X, data_dili$Y!="Less-DILI-Concern")
   Y <- subset(data_dili$Y, data_dili$Y!="Less-DILI-Concern") 
   names(Y)=rownames(X)

   random_seq=sample(nrow(X))
   split_rate=3
   split_sample = suppressWarnings(split(random_seq,1:split_rate))
   Train_X = X[-random_seq[split_sample[[1]]],]
   Train_Y = Y[-random_seq[split_sample[[1]]]]
   Test_X = X[random_seq[split_sample[[1]]],]
   Test_Y = Y[random_seq[split_sample[[1]]]]
   used_model = DF_train(Train_X, Train_Y)

Here is the structure of built Decision Forest Model:

summary(used_model)

Where the performance stored the Fitting performance of training process.
Let's take a look:
- The Accuracy (ACC) is r used_model$performance$ACC;
- The Matttews' Correlation Coefficient (MCC) is r used_model$performance$MCC;
- The Balanced Accuracy (bACC) is r used_model$performance$bACC.

Due to the dataset we used here, we found that DILI is not a well predicted endpoint!

OK, now let's see how about the performance on testing dataset.

Pred_result = DF_pred(used_model,Test_X,Test_Y)

Similarly, let's take a look at the predicting performance:
- The Accuracy (ACC) is r Pred_result$performance$ACC;
- The Matttews' Correlation Coefficient (MCC) is r Pred_result$performance$MCC;
- The Balanced Accuracy (bACC) is r Pred_result$performance$bACC.

Compared the result from training and testing performance:

 matrix(c(round(used_model$performance$ACC,3), round(Pred_result$performance$ACC,3), 
          round(used_model$performance$MCC,3), round(Pred_result$performance$MCC,3), 
          round(used_model$performance$bACC,3), round(Pred_result$performance$bACC,3)),
        nrow=2,ncol=3, dimnames=list(c("Training", "Testing"),c("ACC","MCC","bACC"))
 )

Cross validation analysis

Sometime, you want to run cross validation analysis on the entire dataset.

CV_result = DF_CV(Train_X, Train_Y)
DF_CVsummary(CV_result,plot = T)

Finally, Let's see something Fancy with only one command.

Result = DF_easy(Train_X, Train_Y, Test_X, Test_Y)

seldas/Dforest documentation built on May 30, 2019, 8:08 p.m.