Iris Classification with Scikit-Learn

This project demonstrates a complete Supervised Learning Classification workflow using the Iris Dataset and a Decision Tree Classifier. The objective is to classify Iris flowers into three species based on their physical measurements.

Project Information

Item	Value
Task	Classification
Dataset	Iris Dataset
Samples	150
Features	4
Classes	3
Model	Decision Tree Classifier
Library	Scikit-Learn

Project Workflow

Dataset
↓
Exploratory Data Analysis (EDA)
↓
Feature & Label Selection
↓
Train-Test Split
↓
Model Creation
↓
Model Training
↓
Prediction
↓
Evaluation
↓
Visualization
↓
Conclusion

Dataset

The Iris dataset contains measurements of Iris flowers from three different species:

• Setosa
• Versicolor
• Virginica

Features

• Sepal Length
• Sepal Width
• Petal Length
• Petal Width

Target

• Species

Dataset Source

This dataset was obtained from Kaggle:

• Dataset: Iris Dataset
• Author: Himanshu Nakrani

Exploratory Data Analysis (EDA)

The dataset was analyzed to understand:

• Dataset shape
• Available columns
• Missing values
• Class distribution

The dataset contains:

• 150 samples
• 4 numerical features
• 3 balanced classes

Train-Test Split

The dataset was divided into:

• 80% Training Data
• 20% Testing Data

The parameter stratify=y was used to preserve class distribution in both training and testing sets.

The parameter random_state=42 was used to ensure reproducible results.

Model

A Decision Tree Classifier was used for the classification task.

from sklearn.tree import DecisionTreeClassifier

model = DecisionTreeClassifier()

Evaluation Metrics

The model was evaluated using:

• Accuracy
• Classification Report
• Confusion Matrix

Results

Metric	Value
Accuracy	96.67%
Correct Predictions	29 / 30
Misclassified Samples	1

Visualizations

Confusion Matrix

Shows the number of correct and incorrect predictions for each class.

Scatter Plot

Visualizes the relationship between petal length and petal width across different flower species.

Feature Importance

Displays how much each feature contributes to the model's decision-making process.

Key Findings

• Decision Tree achieved 96.67% accuracy
• Only 1 sample was misclassified
• Petal Width was identified as the most important feature
• The dataset was perfectly balanced
• Setosa was clearly separable from the other species

Future Improvements

Future work may include:

• Random Forest Classifier
• K-Nearest Neighbors (KNN)
• Support Vector Machine (SVM)
• Hyperparameter Tuning
• Model Performance Comparison

Technologies Used

• Python
• Pandas
• Matplotlib
• Scikit-Learn
• Jupyter Notebook

Repository Structure

iris-classification-sklearn/
│
├── data/
│   └── iris.csv
│
├── notebooks/
│   └── iris.ipynb
│
├── images/
│   ├── confusion_matrix.png
│   ├── scatter_plot.png
│   └── feature_importance.png
│
├── README.md
├── requirements.txt
└── .gitignore

Author

This project was created as a beginner machine learning project to practice the complete supervised learning workflow using the Iris dataset.