AI-Text-Analyzer

AI-Text-Analyzer is a powerful Python-based tool designed for Apify to scrape web content and perform deep analysis. It differentiates between human and AI-generated text, checks for grammar errors, and provides detailed readability statistics.

✨ Features

• 🕷️ Smart Scraping: Extracts clean text from any given URL, stripping away ads and scripts.
• 🤖 AI Detection: Uses the roberta-base-openai-detector model to determine if text is Human Written or AI Generated, providing a confidence percentage.
• ✍️ Grammar Checking: Identifies typos and grammar mistakes with context and suggestions (using LanguageTool).
• 📊 Text Statistics: Calculates Flesch Reading Ease, word count, estimated reading time, and sentiment measurement.
• 📝 Persistent Logging: Saves every analysis result as a JSON file in the logs/ directory for historical tracking.

🚀 Installation & Setup

1. Clone the repository:

   git clone <repository-url>
   cd ai-text-analyzer
   

2. Install Dependencies: Note: This includes heavy ML libraries (Torch, Transformers).

   $pip install -r requirements.txt
   

🛠️ Usage

1. Command Line (Direct Input)

The fastest way to test a string of text locally:

$python -m src.main "Your text to analyze goes here..."

Output:

{
  "timestamp": "2024-05-20T10:00:00.123456",
  "ai_detection": {
    "label": "Human Written",
    "confidence_percentage": "98.5%"
  },
  "grammar": [],
  "stats": {
    "word_count": 6,
    "sentiment_polarity": 0.0
  }
}

2. Apify Inputs (Local Simulation)

To simulate an Apify run locally using the configuration file:

1. Edit apify_storage/key_value_stores/default/INPUT.json.
2. Run:

   # Linux/Mac
   export APIFY_LOCAL_STORAGE_DIR="./apify_storage"
   python -m src.main
   
   # Windows Powershell
   $env:APIFY_LOCAL_STORAGE_DIR="./apify_storage"
   python -m src.main
   

💡 Use Cases

| Use Case | Description | | t --- | --- | | Content Moderation | Automatically flag AI-generated spam reviews or comments. | | SEO Optimization | Analyze readability scores (Flesch-Kincaid) to improve search rankings. | | Quality Control | Ensure published articles are free of grammar errors before going live. | | Data Scraping | Harvest text from competitor blogs and analyze their sentiment/tone. |

📂 Project Structure

ai-text-analyzer/
├── apify.json              # Apify Actor configuration
├── Dockerfile              # Container setup (Python 3.11 + Java)
├── requirements.txt        # Dependencies
├── logs/                   # JSON logs of analysis history
├── src/
│   ├── main.py             # Entry point
│   ├── ai_detector.py      # AI detection logic (RoBERTa)
│   ├── grammar_checker.py  # LanguageTool integration
│   ├── scraper.py          # BeautifulSoup web scraper
│   └── text_stats.py       # Statistics & Sentiment

⚠️ Requirements

• Python 3.11+
• Java 8+ (Required for the language-tool-python library)
• Disk Space: ~1GB (for PyTorch and Transformer models)

Ruturaj has created and build this project using AI Suggest more if any additional details can be added.