Student Performance Analysis and Prediction System

A machine learning-based system that analyzes student performance using PAT (Periodic Assessment Tests), SAT (Summative Assessment Tests), and attendance data to provide personalized educational insights and predictions.

Table of Contents

1. Features
2. Technology Stack
3. Project Structure
4. Installation
5. System Architecture
6. Data Structure
7. Machine Learning Implementation
8. API Reference
9. Usage Guide

Features

Core Features

• Performance Prediction: ML-based prediction of student performance in each subject
• Attendance Analysis: Prediction of future attendance patterns and risk assessment
• Class-wise Analytics: Comprehensive analysis of class performance and trends
• Student-wise Analysis: Detailed individual student performance tracking
• Interactive Visualization: Both graphical and tabular data representation
• Personalized Recommendations: AI-driven recommendations for improvement

Key Capabilities

• Real-time performance monitoring
• Subject-wise strength/weakness analysis
• Performance trend visualization
• Risk level assessment
• Data-driven educational recommendations
• Comparative analysis across terms

Technology Stack

Backend

• Python 3.10+
• Flask: Web framework
• scikit-learn: Machine learning implementation
• pandas: Data processing
• numpy: Numerical computations

Frontend

• JavaScript: Interactive UI components
• Chart.js: Data visualization
• HTML5/CSS3: Modern responsive design

Database

• CSV-based data storage (can be extended to SQL databases)

Project Structure

app.py                  # Main Flask application
data/                   # Data directory
├── detailed_student_data.csv   # Primary dataset
├── sample_student_data.csv     # Sample data for testing
└── student_data.csv           # Training data
models/                # Core logic
├── analyzer.py       # Data analysis implementation
└── predictor.py      # ML model implementation
static/               # Frontend assets
├── css/
│   └── style.css    # Application styling
└── js/
    ├── dashboard.js  # Dashboard functionality
    └── main.js      # Core frontend logic
templates/            # HTML templates
├── dashboard.html   # Main dashboard view
└── index.html      # Landing page

Installation

1. Create a virtual environment:

python -m venv venv
.\venv\Scripts\Activate.ps1

2. Install dependencies:

pip install -r requirements.txt

3. Run the application:

python app.py

System Architecture

Core Components

1. StudentAnalyzer (analyzer.py)

Main class responsible for data analysis and insights generation.

Key Methods:

• load_data(): Loads and preprocesses student data
• get_class_list(): Retrieves available classes
• predict_final_marks(student_id): Generates ML-based performance predictions
• predict_future_attendance(student_id): Predicts attendance patterns
• get_student_performance(student_id): Retrieves comprehensive student data
• get_class_insights(class_id): Generates class-level analytics
• identify_weak_subjects(student_id): Analyzes subject-wise performance
• get_recommendations(student_id): Generates personalized recommendations

2. StudentPredictor (predictor.py)

Handles machine learning model training and predictions.

Key Methods:

• train(): Trains ML models for each subject
• predict(): Generates predictions using trained models
• get_recommendations(): Provides ML-based recommendations
• _calculate_confidence(): Determines prediction confidence levels
• _determine_trend(): Analyzes performance trends
• _get_contributing_factors(): Identifies key performance factors

Data Flow

1. Data Loading → Preprocessing → Model Training
2. User Request → Data Analysis → ML Prediction → Result Generation
3. Frontend Display → Interactive Visualization → User Interaction

Data Structure

Student Data Format

student_id,student_name,class,pat_subject_t1,sat_subject_t1,pat_subject_t2,sat_subject_t2,attendance_percentage

Key Fields:

• PAT Scores: Periodic Assessment Test scores (Term 1 & 2)
• SAT Scores: Summative Assessment Test scores (Term 1 & 2)
• Attendance: Student attendance percentage
• Subject Data: Individual scores for English, Math, Science, Social, Computer

Machine Learning Implementation

Model Architecture

• Algorithm: Random Forest Regressor
• Features: Previous test scores, attendance patterns
• Target: Performance prediction per subject
• Validation: Cross-validation with confidence scoring

Prediction Process

1. Data Preparation:

   • Feature scaling using StandardScaler
   • Historical performance analysis
   • Attendance pattern evaluation

2. Model Training:

   • Subject-wise model training
   • Feature importance calculation
   • Confidence score computation

3. Prediction Generation:

   • Performance prediction
   • Trend analysis
   • Contributing factor identification

Confidence Calculation

• Based on prediction variance across trees
• Considers historical data consistency
• Adjusts for attendance patterns

API Reference

Flask Routes

1. GET /student/<student_id>

Returns complete student analysis including:

• Basic information
• Subject-wise performance
• Predictions
• Recommendations

2. GET /class/<class_id>

Returns class-level insights including:

• Performance distribution
• Subject-wise analytics
• Areas of concern

3. POST /upload

Handles data upload and processing

Usage Guide

Dashboard Navigation

1. Class Selection:

   • Choose class from dropdown
   • View class-level insights
   • Access performance distribution

2. Student Analysis:

   • Select student from filtered list
   • Toggle between graph/table views
   • View detailed performance data

3. Performance Tracking:

   • Monitor subject-wise progress
   • View prediction confidence
   • Access contributing factors

Interpreting Results

1. Performance Predictions:

   • Score predictions with confidence levels
   • Trend indicators (Improving/Declining/Stable)
   • Contributing factors analysis

2. Attendance Analysis:

   • Risk level assessment
   • Correlation with performance
   • Future attendance predictions

3. Recommendations:

   • Subject-specific suggestions
   • Priority areas for improvement
   • Attendance-related guidance