Classification and Regression by Random Forest

Prerequisites

In order to run the code in this chapter, you will need to install a number of packages. The packages are listed below.

# Install packages
install.packages("formatR")
install.packages("heplots")
install.packages("ipred")
install.packages("rpart")
install.packages("randomForest")
install.packages("readr")
install.packages("ggplot2")
install.packages("cowplot")
install.packages("mltools")
install.packages("tree")
install.packages("reshape")
install.packages("gridBase")
install.packages("DiagrammeR")
install.packages("caret")

We will also use a dataset in this chapter. The data set contains the expression measure of 4 clinical variables, age and gender for 206 samples. Samples are from people with Alzheimer’s disease (AD), frontotemporal Dementia (FTD), two groups of mild cognitive impairment (MCI), and non-demented controls. We will use both the full dataset and also a limited one (only AD and controls). Here is a small description of the data from the original paper: “This study includes CSF samples from 76 AD patients, 74 mild cognitive impairment (MCI) patients, 11 frontotemporal dementia (FTD) patients, and 45 non-dementia controls. The MCI patients were followed for 4–8 years at 6–12 months intervals and eventually diagnosed with AD (MCI/AD converters) (n=21) or remained at the MCI stage (stable MCI) (n=53)”

Khoonsari, Payam Emami et al. ‘Improved Differential Diagnosis of Alzheimer’s Disease by Integrating ELISA and Mass Spectrometry-Based Cerebrospinal Fluid Biomarkers’. 1 Jan. 2019 : 639 – 651.