🎓 EduVision AI

AI-Enhanced Student Performance Prediction and Management System

An intelligent desktop application designed to monitor academic performance, predict at-risk students, and provide academic intervention recommendations using Machine Learning.

---

📑 Table of Contents

• 📌 Project Overview
• ✨ Core Features
• 🤖 AI Features
• 🛠️ Technologies Used
• 🖼 User Interface
• 🚀 Installation & Setup
• 🔄 Rebuild Database & AI Model
• ⚙️ How the System Works
• 👨‍💻 Developer

---

📌 Project Overview

EduVision AI is a desktop-based student management system that combines traditional academic management with Artificial Intelligence to support academic performance monitoring and early intervention.

The system enables teachers and administrators to manage students, classes, assessments, and reports while utilizing Machine Learning models to identify at-risk students and generate personalized recommendations.

Objectives

• Monitor academic performance
• Predict student risk levels
• Provide AI-powered recommendations
• Support early academic intervention
• Analyze classroom performance
• Improve educational decision-making

---

✨ Core Features

👑 Admin Module

Full system access:

• Secure login system
• Manage teachers
• Manage classes
• Assign teachers to classes
• Monitor student records
• Add, edit, and delete students
• System overview dashboard

👨‍🏫 Teacher Module

Class-level access:

• Secure login system
• Manage assigned classes
• Record assessment scores
• Generate AI predictions
• Export Class Inforamtion to PDF and Excel
• Export reports to PDF
• View classroom analytics

🎓 Student Prediction Sandbox

Prediction-only access:

• No login required
• Enter assessment scores manually
• Receive instant AI predictions
• View personalized recommendations
• Export prediction report as PDF

Student Sandbox Mode is used for demonstration and self-analysis purposes. Data entered here is not stored in the database.

---

🤖 AI Features

EduVision AI uses Machine Learning techniques to evaluate student performance based on multiple academic indicators.

Prediction Factors

• Attendance
• Quiz
• Homework
• Assignment
• Midterm Exam
• Final Exam
• Participation
• Project
• Behavior

AI Outputs

• Predicted Academic Score
• Risk Classification
• Performance Analysis
• Personalized Recommendations

Risk Levels

Risk Level	Description
🟢 Low Risk	Student is performing well
🟡 Medium Risk	Student requires monitoring
🔴 High Risk	Student requires intervention

---

🛠️ Technologies Used

Technology	Purpose
Python 3.11	Core Programming Language
CustomTkinter	Modern Desktop GUI
SQLite	Local Database
Pandas	Data Processing
NumPy	Numerical Computation
Scikit-Learn	Machine Learning
Matplotlib	Data Visualization
ReportLab	PDF Report Generation
OpenPyXL	Excel Export

---

🖼User Interface

---

🚀 Installation & Setup

⚠️ Before starting, make sure you have Python 3.11.X installed. Most of the libraries and dependencies used in this project were developed and tested with Python 3.11, and using other versions may cause installation or compatibility issues.

1. Clone Repository

Open VS Code Terminal or Command Prompt:

git clone https://github.com/Laykimhoung/student-performance-prediction.git
cd student-performance-prediction

---

2. Create Virtual Environment

Open VS Code Terminal and run:

py -3.11 -m venv .venv

This creates an isolated Python environment for the project.

---

3. Activate Virtual Environment

In VS Code Terminal and run:

Windows:

.venv\Scripts\activate

Mac/Linux:

source .venv/bin/activate

If successful, you should see:

(.venv) PS C:\YourProject\student-performance-prediction>

✅ This means the virtual environment is active.

---

⚠️ First-Time PowerShell Users

If you receive an error similar to:

running scripts is disabled on this system

You must allow PowerShell scripts to run.

Step A — Open PowerShell as Administrator

1. Search for PowerShell
2. Right-click Windows PowerShell
3. Select Run as Administrator

Run:

Set-ExecutionPolicy RemoteSigned -Scope CurrentUser

When prompted:

Do you want to change the execution policy?
[Y] Yes

Type:

Y

and press Enter.

---

Step B — Activate Environment

Return to the VS Code Terminal and run:

.venv\Scripts\Activate.ps1

If successful, you should see:

(.venv) PS C:\YourProject\student-performance-prediction>

✅ This means the virtual environment is active.

---

4. Install Dependencies

With the virtual environment activated, run in the VS Code Terminal:

pip install -r requirements.txt

⏳ Please Be Patient

The installation may appear stuck while installing large packages. Depending on your internet speed and computer performance, installation may take:

5–10 minutes

Do NOT close the terminal.

Wait until you see:

Successfully installed ...

before proceeding.

---

🔧 Troubleshooting Installation Issues

Problem 1: Installation Failed

If package installation fails or the application reports missing modules such as:

No module named 'customtkinter'

or

No module named 'PIL'

recreate the virtual environment.

Step 1 — Deactivate Environment

Run in VS Code Terminal:

deactivate

Step 2 — Delete Old Virtual Environment

Remove-Item -Recurse -Force .venv

Step 3 — Create New Virtual Environment

py -3.11 -m venv .venv

Step 4 — Activate Environment

.venv\Scripts\Activate.ps1

Step 5 — Reinstall Dependencies

pip install -r requirements.txt

Wait until installation finishes completely.

---

5. Run Application

Run this to start EduVision AI:

python app.py

If everything is installed correctly, EduVision AI will launch successfully.

The repository already includes the database and trained AI models. Most users only need to install the requirements and run the application.

---

🔄 Rebuild Database & AI Model

If you want to generate a fresh demo database and retrain the AI model from scratch, follow these steps:

Step 1 — Delete Database

Delete:

database/eduvision.db

Step 2 — Recreate Database Tables

python -m database.schema

Step 3 — Generate Demo Data

python -m database.seed

Step 4 — Retrain AI Models

python -m ai.trainer

This will automatically overwrite the existing AI model files:

ai/model/risk_model.pkl
ai/model/score_model.pkl

After training is completed, simply run:

python app.py

to start EduVision AI with the newly generated database and AI models.

---

⚙️ How the System Works

Step 1 — Data Collection

Teachers enter student assessment scores into the system.

Step 2 — Academic Evaluation

The system calculates student performance metrics and academic indicators.

Step 3 — AI Prediction

The trained Machine Learning model analyzes student performance patterns and predicts:

• Academic Score
• Risk Level
• Intervention Priority

Step 4 — Recommendation Generation

The recommendation engine generates personalized improvement suggestions based on student weaknesses.

Step 5 — Reporting

Users can export performance reports and prediction results in PDF format.

---

👨‍💻 Developer

Lay Kimhoung (Xakuraii)

Computer Science Student

Project: EduVision AI Version: 1.0 Year: 2026

---

Academic Research Topic

Artificial Intelligence in Student Management Systems to Enhance Academic Performance Monitoring and Intervention