Academic Text Analyzer

An AI-powered web application that analyzes text authenticity using advanced detection, featuring OCR capabilities for image-to-text, analysis modes, and history

Comment

Inspiration

The rise of AI-generated content has made it increasingly difficult to distinguish between human-written and machine-generated text. This project was inspired by the need for a reliable, user-friendly tool that can help educators, content creators, and professionals verify text authenticity in an era where AI writing tools are becoming ubiquitous.

What it does

The Text Authenticity Detector provides:

  • Text Analysis: Advanced AI detection algorithms that analyze text patterns to identify potentially AI-generated content
  • OCR Integration: Upload images containing text and extract it automatically using Tesseract.js
  • Dual Analysis Modes: Toggle between "Critical" mode (strict detection) and "Generous" mode (fair assessment)
  • History Tracking: Local storage-based history system to review previous analyses
  • Responsive Design: Modern, accessible interface built with shadcn/ui components

How we built it

  • Frontend Framework: React with TypeScript for type safety and component-based architecture
  • Styling: Tailwind CSS with shadcn/ui component library for consistent, modern design
  • OCR Technology: Tesseract.js for optical character recognition from uploaded images
  • State Management: React hooks and local storage for data persistence
  • Build Tool: Vite for fast development and optimized production builds
  • UI Components: Custom components built on Radix UI primitives for accessibility

Challenges we ran into

  • OCR Accuracy: Fine-tuning Tesseract.js settings to achieve reliable text extraction from various image qualities
  • Analysis Algorithm: Balancing detection sensitivity between the critical and generous modes
  • Local Storage Management: Implementing efficient history storage without overwhelming browser storage limits
  • Responsive Design: Ensuring the interface works seamlessly across different screen sizes and devices

Accomplishments that we're proud of

  • Successfully integrated OCR functionality with seamless user experience
  • Created a flexible analysis system with customizable detection modes
  • Implemented a comprehensive history system using local storage
  • Built a fully responsive, accessible interface following modern design principles
  • Achieved smooth text processing workflow from input to analysis results

What we learned

  • Advanced integration of OCR libraries in web applications
  • Effective state management patterns for complex user interactions
  • Best practices for local storage implementation and data persistence
  • Modern React development patterns with TypeScript
  • Accessibility considerations in UI component design

Project Documentation

Technologies Used

  • Frontend: React 18, TypeScript, Vite
  • Styling: Tailwind CSS, shadcn/ui
  • OCR: Tesseract.js
  • State Management: React Hooks, Local Storage
  • Icons: Lucide React
  • Build Tools: Vite, TypeScript compiler # OCR processing utilities

Features

  1. Text Input: Manual text entry with scrollable textarea
  2. Image Upload: OCR processing with loading states and error handling
  3. Analysis Modes: Toggle between critical and generous detection
  4. Results Display: Comprehensive analysis with confidence scores
  5. History System: Local storage-based previous analysis tracking
  6. Responsive Design: Mobile-first approach with desktop optimization

Built With

Share this project:

Updates