Rahul Fernandes 2023-09-08
The IRIS Classification Using R repository conducts exploratory data analysis on the IRIS dataset and a Polynomial Support Vector Machine classification model is built to classify the three species of the Iris flower.
library(datasets)
library(dplyr)
library(ggplot2)
library(GGally)
library(caret)Summary of Statistics for the whole Dataset
## Sepal.Length Sepal.Width Petal.Length Petal.Width
## Min. :4.300 Min. :2.000 Min. :1.000 Min. :0.100
## 1st Qu.:5.100 1st Qu.:2.800 1st Qu.:1.600 1st Qu.:0.300
## Median :5.800 Median :3.000 Median :4.350 Median :1.300
## Mean :5.843 Mean :3.057 Mean :3.758 Mean :1.199
## 3rd Qu.:6.400 3rd Qu.:3.300 3rd Qu.:5.100 3rd Qu.:1.800
## Max. :7.900 Max. :4.400 Max. :6.900 Max. :2.500
## Species
## setosa :50
## versicolor:50
## virginica :50
##
##
##
Descriptive statistics for Sepal Length by Species
## # A tibble: 3 × 3
## Species `mean(Sepal.Length)` `sd(Sepal.Length)`
## <fct> <dbl> <dbl>
## 1 setosa 5.01 0.352
## 2 versicolor 5.94 0.516
## 3 virginica 6.59 0.636
Descriptive statistics for Sepal Width by Species
## # A tibble: 3 × 3
## Species `mean(Sepal.Width)` `sd(Sepal.Width)`
## <fct> <dbl> <dbl>
## 1 setosa 3.43 0.379
## 2 versicolor 2.77 0.314
## 3 virginica 2.97 0.322
Descriptive statistics for Petal Length by Species
## # A tibble: 3 × 3
## Species `mean(Petal.Length)` `sd(Petal.Length)`
## <fct> <dbl> <dbl>
## 1 setosa 1.46 0.174
## 2 versicolor 4.26 0.470
## 3 virginica 5.55 0.552
Descriptive statistics for Petal Width by Species
## # A tibble: 3 × 3
## Species `mean(Petal.Width)` `sd(Petal.Width)`
## <fct> <dbl> <dbl>
## 1 setosa 0.246 0.105
## 2 versicolor 1.33 0.198
## 3 virginica 2.03 0.275
From the above, the dataset has three species of Iris flowers, Setosa, Versicolor and Virginica. There are 50 samples of each of these species, hence the dataset is balanced. In terms of Sepal Length, Virginica has the highest mean making it generally bigger on average with respect to Sepal Length. In terms of Sepal Width, Setosa has the highest mean making it generally bigger on average with respect to Sepal Width. In terms of Petal Width and Length, Virginica has the highest mean making it roughly the biggest species among the three.
There is an overlap in Sepal features among the three Species and there is no overlap in Petal features among the three species.
From the above pair plot, there is almost no overlap between features of the Setosa species, whereas there is overlap over features between Versicolor and Virginica species.
It can be inferred that Virginica is generally the biggest among the three species of the Iris flower and Setosa is the smallest. The relationship between the features and species shows that there is an overlap of features between Versicolor and Virginica. The mean and standard deviation of the Sepal Width feature are similar among the three species. This might indicate that this feature might not be as important while trying to differentiate between the three species. The graph of relationship between features by species shows that the classification is not linear.
The dataset is split into two parts, training and testing, 80% of the dataset is used for training and the rest 20% is used for testing.
set.seed(100)
TrainingIndex <- createDataPartition(iris$Species, p=0.8, list = FALSE)
TrainingSet <- iris[TrainingIndex,]
TestingSet <- iris[-TrainingIndex,]# Build Training Model
svmModel <- train(Species ~ ., data = TrainingSet,
method = "svmPoly",
na.action = na.omit,
preProcess=c("scale","center"),
trControl= trainControl(method="none"),
tuneGrid = data.frame(degree=1,scale=1,C=1)
)
print(svmModel)## Support Vector Machines with Polynomial Kernel
##
## 120 samples
## 4 predictor
## 3 classes: 'setosa', 'versicolor', 'virginica'
##
## Pre-processing: scaled (4), centered (4)
## Resampling: None
# Model performance
svmModel.testing <- predict(svmModel, TestingSet)
svmModel.testing.confusion <- confusionMatrix(svmModel.testing,
TestingSet$Species
)
# Print Confusion Matrix
print(svmModel.testing.confusion)## Confusion Matrix and Statistics
##
## Reference
## Prediction setosa versicolor virginica
## setosa 10 0 0
## versicolor 0 9 0
## virginica 0 1 10
##
## Overall Statistics
##
## Accuracy : 0.9667
## 95% CI : (0.8278, 0.9992)
## No Information Rate : 0.3333
## P-Value [Acc > NIR] : 2.963e-13
##
## Kappa : 0.95
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: setosa Class: versicolor Class: virginica
## Sensitivity 1.0000 0.9000 1.0000
## Specificity 1.0000 1.0000 0.9500
## Pos Pred Value 1.0000 1.0000 0.9091
## Neg Pred Value 1.0000 0.9524 1.0000
## Prevalence 0.3333 0.3333 0.3333
## Detection Rate 0.3333 0.3000 0.3333
## Detection Prevalence 0.3333 0.3000 0.3667
## Balanced Accuracy 1.0000 0.9500 0.9750
From the above graph, the feature Sepal Width has the least importance in the model when classifying the three species of the Iris flower. Hence, proving the hypothesis made during analysis of the dataset.
A Polynomial Support Vector Machine classification model is trained on 80% of the IRIS Dataset and tested on 20% of the IRIS Dataset. The model was compared with KNN and Random Forest models and the SVM model showed the highest metrics among the other models with an accuracy of 96.67% in classifying the three species of the Iris flower.