🎓 Student Performance Prediction using Azure Machine Learning Designer

End-to-End Machine Learning Pipeline for Classification & Regression

---

📖 Project Overview

This project demonstrates an end-to-end Machine Learning pipeline built using Microsoft Azure Machine Learning Designer.

The pipeline combines Binary Classification and Regression models in a single workflow to predict:

• 🎯 Student Placement Status
• 📊 Student Exam Score

The project covers the complete ML lifecycle including:

• Data Preparation
• Feature Selection
• Missing Value Handling
• Model Training
• Model Evaluation
• Performance Comparison

---

🎯 Objectives

✅ Predict whether a student will be placed

✅ Predict a student's final exam score

✅ Compare Classification and Regression models

✅ Build a complete Azure ML Designer Pipeline

---

📂 Dataset

The dataset contains student academic and demographic information.

Target Variables

Task	Target Variable
Classification	placement_status
Regression	exam_score

Example Features

• Age
• Gender
• Attendance
• Study Hours
• Previous Grades
• Assignments
• Family Income
• Internet Access
• Sleep Hours
• Motivation Level
• Teacher Rating
• Extracurricular Activities

---

🏗 Azure ML Pipeline

Dataset
   │
   ▼
Summarize Data
   │
   ▼
Select Columns
   │
   ▼
Clean Missing Data
   │
   ▼
Split Data
   │
   ├──────────────► Logistic Regression
   │                     │
   │                 Train Model
   │                     │
   │                 Score Model
   │                     │
   │               Evaluate Model
   │
   └──────────────► Linear Regression
                         │
                     Train Model
                         │
                     Score Model
                         │
                   Evaluate Model

---

🧠 Machine Learning Models

Model	Task
Two-Class Logistic Regression	Binary Classification
Linear Regression	Regression

---

📊 Classification Results

Metric	Value
Accuracy	98.36%
AUC	0.9988
Precision	98.61%
Recall	99.42%
F1 Score	99.02%

---

📈 Interpretation

Metric	Meaning
Accuracy	Correct predictions made by the model
AUC	Ability to distinguish between classes
Precision	Percentage of predicted positives that are correct
Recall	Percentage of actual positives correctly identified
F1 Score	Balance between Precision and Recall

---

📉 Regression Results

Metric	Value
R² Score	0.8110
Mean Absolute Error	5.32
RMSE	6.59
Relative Absolute Error	0.416
Relative Squared Error	0.189

---

📈 Interpretation

Metric	Meaning
R² Score	Explains 81% of the variation in exam scores
MAE	Average prediction error
RMSE	Penalizes larger prediction errors
Relative Errors	Lower values indicate better model performance

---

📸 Project Screenshots

Azure ML Designer Pipeline

---

Classification Metrics

---

Regression Metrics

---

🚀 Technologies Used

Category	Tools
Cloud Platform	Microsoft Azure
ML Platform	Azure Machine Learning Designer
Algorithms	Logistic Regression, Linear Regression
ML Tasks	Classification, Regression
Data Processing	Data Cleaning, Feature Selection
Evaluation	Accuracy, AUC, Precision, Recall, F1, RMSE, R²

---

📁 Repository Structure

Student-Performance-Prediction-AzureML
│
├── README.md
├── LICENSE
├── dataset
│   └── student_dataset_10000_rows.csv
│
├── report
│   └── Project_Report.pdf
│
└── screenshots
    ├── pipeline.png
    ├── classification_metrics.png
    ├── regression_metrics.png
    ├── roc_curve.png
    └── pr_curve.png

---

🎓 Key Learning Outcomes

✔ Azure Machine Learning Designer

✔ Data Preprocessing

✔ Feature Engineering

✔ Binary Classification

✔ Regression

✔ Model Evaluation

✔ Performance Metrics

✔ End-to-End ML Pipeline

---

📌 Future Improvements

• Deploy the trained model as a REST API
• Hyperparameter tuning
• Try Decision Tree and Random Forest models
• Implement Automated ML for comparison
• Build a Power BI dashboard for predictions

---

🤝 Connect with Me

Osama Ghafoor

💼 AI | Machine Learning | Azure AI | Cloud Computing

📧 osamag13@gmail.com

🔗 LinkedIn: www.linkedin.com/in/osama-ghafoor-b87346a1

---

⭐ If you found this project helpful...

Please consider giving it a Star ⭐

It motivates me to build and share more AI & Machine Learning projects.

---

🚀 Built with Microsoft Azure Machine Learning Designer

Classification • Regression • Azure AI • Machine Learning