Data Science Desktop Survival Guide by Graham Williams
Desktop Survival Project Home Preface Data Science Introducing R R Constructs R Tasks R Strings R Read, Write, and Create Data Template Data Exploration Data Wrangling Data Visualisation Statistics ML Template ML Scenarios ML Activities ML Applications ML Algorithms Cluster Analysis Decision Trees Computer Vision Graph Data Privacy Literate Data Science Coding with Style Resources Bibliography Index

CLICK HERE TO VISIT THE UPDATED SURVIVAL GUIDE

Decision Trees Setup

20180603 Packages used in this chapter include C50, RWeka, party, partykit, rpart, rpart.plot, and rattle.

Packages are loaded into the currently running R session from your local library directories on disk. Missing packages can be installed using utils::install.packages() within R. On Ubuntu, for example, R packages can be installed using wajig install r-cran-<pkgname>.

# Load required packages from local library into the R session.

library(C50)          # Original C5.0 implementation.
library(RWeka)        # Weka decision tree J48.
library(party)        # Conditional decision trees ctree().
library(partykit)     # Convert rpart object to BinaryTree
library(rattle)       # GUI for building trees and fancy tree plot.
library(rpart)        # Popular decision tree algorithm.
library(rpart.plot)   # Enhanced tree plots.

The rattle::weatherAUS dataset is loaded into the template variable ds and further template variables are setup as introduced in Williams (2017). See Chapter 7 for details.

dsname <- "weatherAUS"
ds     <- get(dsname)

nobs   <- nrow(ds)

vnames <- names(ds)
ds    %<>% clean_names(numerals="right")
names(vnames) <- names(ds)

vars   <- names(ds)
target <- "rain_tomorrow"
vars   <- c(target, vars) %>% unique() %>% rev()

It is always useful to remind ourselves of the dataset with a random sample:

ds  %>% sample_frac() %>% select(date, location, sample(3:length(vars), 5))

## # A tibble: 176,747 x 7
##    date       location max_temp rain_today humidity_9am wind_dir_9am
##    <date>     <chr>       <dbl> <fct>             <int> <ord>       
##  1 2010-09-03 Adelaide     18.3 No                   67 ENE         
##  2 2010-10-30 WaggaWa~     19.3 No                   81 NE          
##  3 2017-04-04 GoldCoa~     25.5 No                   65 S           
##  4 2018-10-28 Witchcl~     24.7 No                   79 SE          
....