🎓 Student Dropout Prediction (Machine Learning Project)

📌 Project Overview

This project builds a machine learning model to predict whether a student is likely to drop out. Early identification of at-risk students helps institutions take preventive actions and improve retention.

---

📂 Project Structure

Student-Dropout-ml-project
├── data/
├── images/
│   ├── confusion_matrix.png
│   ├── roc_curve.png
│   ├── precision_recall_curve.png
├── src/
│   └── student_dropout.py
├── notebook/
│   └── student_dropout.ipynb
├── models/
├── README.md

---

⚙️ Tech Stack

• Python
• NumPy
• Pandas
• Matplotlib
• Seaborn
• Scikit-learn

---

🧠 ML Pipeline

1. Data Preprocessing

• Dropped irrelevant column: Student_ID
• Removed missing values using dropna()
• Split dataset into train/test (80/20)

2. Feature Engineering

• Categorical Features → OneHotEncoder
• Numerical Features → StandardScaler
• Skewed Features → PowerTransformer (Yeo-Johnson)

3. Pipeline

Used ColumnTransformer + Pipeline to ensure consistent preprocessing and modeling.

---

🤖 Model Used

• Logistic Regression

• Parameters:

  • solver = "lbfgs"
  • max_iter = 1000
  • class_weight = "balanced"
  • random_state = 42

---

📊 Model Performance

Classification Report

Accuracy: 0.75

Class 0 (Non-Dropout):
Precision: 0.90
Recall: 0.75
F1-score: 0.82

Class 1 (Dropout):
Precision: 0.48
Recall: 0.74
F1-score: 0.58

---

📉 Confusion Matrix

---

📈 ROC Curve (AUC ≈ 0.82)

---

📈 Precision-Recall Curve

---

🔍 Key Insights

• Model achieves 75% accuracy
• Strong recall (0.74) for dropout class → good at identifying at-risk students
• Lower precision (0.48) → more false positives

---

🚀 Future Improvements

• Improve precision using:

  • Advanced models (Random Forest, XGBoost)
  • Hyperparameter tuning
  • Better feature engineering

• Handle class imbalance with SMOTE / SMOTENC

• Deploy model using Streamlit or Flask

---

▶️ How to Run

1. Clone the repository

git clone https://github.com/<your-username>/Student-Dropout-ml-project.git
cd Student-Dropout-ml-project

2. Install dependencies

pip install -r requirements.txt

3. Run the script

python src/student_dropout.py

---

📌 Notes

• Dataset is not included (add your own CSV file)
• Ensure file name: student_dropout_dataset_v3.csv

---

👨‍💻 Author

Durgesh Swarnakar