Apppublic

sanyab/SEO-Intelligence-Tool

sourceHugging Facemitupdated 1y agoView on Hugging Face

0likes

App README

📘 SEO Content Intelligence Tool

An AI-powered web application that analyzes web articles and blog content to generate SEO insights — including keyword extraction, topic detection, and meta tag suggestions.

📌 Problem Statement

In the digital marketing and content publishing space, optimizing content for search engines is critical for visibility. However, identifying relevant keywords, extracting topics, and generating effective meta titles/descriptions is often time-consuming and manual.

Challenge: Build a lightweight, deployable SEO tool that can automate this process using modern NLP techniques.

✅ Solution Overview

This tool enables users to extract SEO-relevant insights from:

—Raw text content
—Uploaded documents (.txt, .pdf, .doc, .docx)
—Blog/article URLs (via web scraping)

It performs the following:

—Keyword extraction using KeyBERT (transformer-based)
—Named Entity Recognition (NER) and noun chunk detection via spaCy
—SEO meta title and description generation
—Support for OCR-based scanned PDFs using Tesseract
—Unified interface built with Streamlit, deployed on Hugging Face Spaces

🛠️ How It Works

👇 Input Options

—Raw Text: Direct paste input
—File Upload: Accepts .txt, .pdf, .doc, .docx
—URL Input: Scrapes article content using BeautifulSoup

🧠 AI-Powered Analysis

Feature	Description
Top Keywords	Extracted via `KeyBERT` using transformer embeddings
Named Entities	Extracted via `spaCy`
Topics / Noun Phrases	Based on noun chunks from spaCy’s parser
Meta Title & Description	Heuristically generated from top-ranked keywords and leading sentences

🖼️ Application Demo

🔗 Live Demo: https://huggingface.co/spaces/your-username/seo-intelligence-tool

Demo Example Video

📁 Project Structure

bash

.
├── src
|     ├── app.py           # Streamlit UI and main controller
|     ├── nlp_utils.py     # NLP functions for keyword, entity, and meta extraction
|     ├── file_utils.py    # File/URL scraping and OCR handling
├── requirements.txt       # Python package dependencies
├── .huggingface.yml       # Config for installing system packages (OCR support)
└── README.md              # Project documentation

⚙️ Installation & Local Setup

bash

# Clone the repository
git clone https://github.com/sanyab1801/seo-intelligence-tool.git
cd seo-intelligence-tool

# Install Python dependencies
pip install -r requirements.txt

# Optional: install OCR dependencies (for .pdf scans, .doc support)
sudo apt install tesseract-ocr poppler-utils

# Run the app
streamlit run app.py

📝 On Windows, install Tesseract OCR and Poppler manually, and add to PATH.

📦 Requirements

txt

streamlit
spacy
keybert
sentence-transformers
PyPDF2
python-docx
textract
pytesseract
pdf2image
beautifulsoup4
requests

The app uses the en_core_web_sm spaCy model. It will auto-download on first run.

📂 Supported File Types

Format	Support	Notes
`.txt`	✅	Direct text read
`.pdf`	✅	Uses PyPDF2 or Tesseract OCR if scanned
`.docx`	✅	Parsed using `python-docx`
`.doc`	✅	Extracted using `textract`
Scanned `.pdf`	✅	OCR performed via `pytesseract`

🧠 Future Improvements (Optional Enhancements)

—✅ Export results to .csv or .json
—✅ Keyword density and readability scoring
—✅ SEO similarity comparison with competitor URLs
—✅ Language detection and support for multilingual input

👨‍💻 Author

Developed by \[Sanya Behera] GitHub: github.com/sanyab1801 Deployed on: Hugging Face Spaces