📊 Student Habits & Performance Analytics System

Group 2 — Team 07 | CADT IDT · Second Year · OOP (Python)
Lecturer: Han Leangsiv | Seat Sreylenn · Chhim SokSambath · Yin Sousdey
Academic Year 2024–2025

---

1. Project Overview

The Student Habits & Performance Analytics System is a Python-based desktop application that analyzes student behavioral data and predicts exam performance using Machine Learning. It is built with Object-Oriented Programming principles and presented through a modern dark-themed CustomTkinter GUI no command-line interaction required.

The system loads data from a real-world dataset of 1,000 students (sampled from 80,000) containing 30+ behavioral, health, socio-economic, and academic attributes per student. A Random Forest Regressor (scikit-learn) is trained on 24 behavioral features to predict each student's exam_score (0–100).

Key Functionality

• Analytics — View all students, top performers, at-risk students, dropout-risk flags, score distribution charts, semester averages, and major rankings
• Prediction — Predict a student's exam score by entering a manual behavioral profile or by looking up an existing student ID
• OOP Architecture — Implements Encapsulation, Inheritance, Polymorphism, Abstraction, and 5 Magic Methods across the Student, UndergraduateStudent, and GraduateStudent class hierarchy
• Dual Data Source — Supports both CSV file loading and MySQL database connection
• Report Export — Generates and saves individual student .txt reports to data/reports/

---

2. Steps to Run the Code

Step 1 — Clone or extract the project

# If using Git
git clone https://github.com/sslenn/Cadtalytics
cd Cadtalytics

# If using the ZIP file
# Extract the ZIP, then navigate into the proj2_out/ folder

Step 2 — Install dependencies

pip install -r requirements.txt

Step 3 — Run the application

python main.py

The app will:

1. Automatically load data/students_habits.csv
2. Build all student objects
3. Train the Random Forest model in the background (~2–5 seconds)
4. Display the GUI with all 13 views ready to use

Note: The Prediction views will show a loading message until the RF model finishes training. All Analytics views are available immediately.

Step 4 — (Optional) Load from MySQL instead of CSV

If you have MySQL running, first import the database:

mysql -u root -p < students_1000.sql

Then when the app starts, select option 2 (Load from MySQL database). Update the connection settings in main.py if needed:

FileHandlerMySQL(
    host="127.0.0.1",
    port=8889,
    user="root",
    password="root",
    database=students_1000"
)

---

3. Dependencies & Installation Instructions

Requirements

Library	Version	Purpose
customtkinter	>= 5.2.0	Modern dark-themed GUI framework
matplotlib	>= 3.7.0	Embedded charts and visualizations
numpy	>= 1.24.0	Array operations for ML
scikit-learn	>= 1.3.0	RandomForestRegressor, train_test_split, R², MAE
mysql-connector-python	>= 8.0.0	Optional MySQL data source

Installation

All dependencies can be installed with a single command:

pip install customtkinter matplotlib numpy scikit-learn mysql-connector-python

Or using the provided requirements file:

pip install -r requirement.txt

Python Version

• Python 3.10 or higher is required
• Check your version: python --version
• Download from: https://www.python.org/downloads/

Operating System

Compatible with Windows, macOS, and Linux.

---

4. GitHub Link

https://github.com/sslenn/Cadtalytics

---

Project Structure

Cadtalytics/
│
├── main.py                     # GUI entry point
├── requirement.txt             # Python dependencies
├── README.md                   # This file
│
├── data/
│   ├── students_habits.csv     # 1,000-student dataset
│   └── reports/                # Exported .txt and .png reports
│
└── src/
    ├── student.py              # Base Student class
    ├── student_types.py        # UndergraduateStudent & GraduateStudent
    ├── grade_manager.py        # Manages Student objects
    ├── analytics.py            # Statistical analysis
    ├── visualizer.py           # Chart generation
    ├── predictor.py            # Random Forest model
    ├── file_handler.py         # CSV read/write/export
    └── file_handler_mysql.py   # MySQL connection

---

Cambodia Academy of Digital Technology (CADT) — OOP (Python) Final Project