Prediction of species from morphology from iris dataset

Parikshit Sanyal
09 Sep 18

Introduction

The iris dataset contains the following parameters to describe the flower of a particular species

data(iris)
names(iris)

[1] "Sepal.Length" "Sepal.Width"  "Petal.Length" "Petal.Width" 
[5] "Species"

The application

This applications inputs the four parameters as slider values, and makes a scatter plot

plot(c(1,2,3,4),ylab='Params',pch=19,col='red')

plot of chunk unnamed-chunk-2

Prediction model

The model is trained by the rpart method

inTrain <- createDataPartition(y=iris$Species,p=0.75,list=FALSE)
tr <- iris[inTrain,]
ts <- iris[-inTrain,]
mod_rpart <- train(Species ~ .,tr,method='rpart')

Predicting species

The outputs is generated by taking the inputs and passing them as a dataframe

output$species <- reactive(predict(mod_rpart,data.frame(Sepal.Length=input$Sepal.Length,                                                 Sepal.Width=input$Sepal.Width,                                           Petal.Length=input$Petal.Length,
Petal.Width=input$Petal.Width)))
})