Student Assessment System

Project Information

Developers

• Shoun Salaji (2447248)
• Yojit Shinde (2447260)

Guide

Dr. Sudhakar Tharuman
Department of Computer Science
CHRIST (Deemed to be University), Bengaluru-29

Table of Contents

1. Overview
2. Features
3. Setup
4. Usage
5. Architecture
6. Technology Stack
7. Screenshots
8. Development Status
9. SDG Alignment

Overview

The Student Assessment System is an advanced grading solution designed to automate evaluation processes, detect plagiarism, and identify malpractice in academic submissions. It leverages machine learning, AI-based content analysis, and OCR for handwritten assignments to ensure academic integrity and streamline grading workflows.

Features

• Answer Verification

  • Automated evaluation against predefined answer keys
  • Plagiarism detection
  • Malpractice identification

• Assignment Verification

  • Originality checking
  • Structure and format validation
  • Peer-to-peer copying detection

• Advanced Analysis

  • Online plagiarism detection
  • AI-generated content identification
  • Handwritten text OCR conversion

Setup

Prerequisites

• Python 3.8 or higher
• Google Cloud Platform account
• Internet connection for API access

Installation Steps

1. Clone Repository

   git clone https://github.com/sho6000/Student-Assessment-Sys.git
   cd Student-Assessment-Sys

2. Install Dependencies

   pip install -r requirements.txt

3. Google Cloud Vision Setup

   1. Create a Google Cloud Project
   2. Enable the Cloud Vision API
   3. Create a service account and download credentials
   4. Rename credentials to key.json
   5. Place in project root directory

   Note: key.json is git-ignored for security

4. Test Files Setup

   • Create test_files directory structure:

     test_files/
     ├── assignments/
     ├── answer_sheets/
     └── sample_documents/
     

   Note: test_files directory is git-ignored

Usage

1. Start Application

   streamlit run app.py

2. Available Operations

   • Upload and verify answer sheets
   • Process assignments for plagiarism
   • Detect AI-generated content
   • Convert handwritten documents to text

Architecture

Frontend

• Streamlit-based web interface
• Intuitive user controls
• Real-time feedback

Backend

• Plagiarism detection engine
• Malpractice detection module
• Answer key matching algorithms

APIs and Services

• Google Cloud Vision for OCR
• Google Gemini for content analysis
• Custom ML models for verification

Technology Stack

• Core: Python
• Frontend: Streamlit
• APIs:
  • Google Cloud Vision API
  • Google Gemini API

Screenshots

Home Screen

Answer Verification

Assignment Verification

OCR Results

Development Status

In Progress

• OCR accuracy optimization
• Enhanced malpractice detection
• Refined AI content analysis

Planned Features

• Multi-language support
• Batch processing capabilities
• Advanced analytics dashboard

SDG Alignment

SDG 4: Quality Education

• Fair and accurate assessments
• Transparent learning outcomes
• Enhanced educational integrity

SDG 16: Peace, Justice, and Strong Institutions

• Academic integrity reinforcement
• Ethical practice promotion
• Transparent evaluation systems