entity-extractor

Entity Extraction App

A multi-page Streamlit app for extracting named entities from various sources using the Dandelion.eu API.

Features

Three extraction modes:

1. SERPs

• Search Google for a keyword
• Extract content from top-ranking pages
• Identify common entities across the SERP
• Visualize entity frequency

2. CSV

• Upload keyword/URL exports
• Extract entities from grouped keywords
• Batch process multiple URLs

3. YouTube

• Paste a YouTube URL
• Extract entities from video transcript
• Identify key topics discussed

Requirements

pip install -r requirements.txt

API Keys Required

Dandelion.eu (Required for all pages)

• Free tier: 1,000 requests/day
• Sign up at dandelion.eu

ValueSERP (Required for SERPs page only)

• PAYG pricing, no expiration
• Sign up at valueserp.com

Usage

1. Start the app:

   streamlit run Home.py

2. Enter your API keys in the sidebar

3. Select a page from the sidebar:

   • SERPs - Search and extract from Google results
   • CSV - Upload a keyword file
   • YouTube - Paste a video URL

4. Download results as CSV

Configuration Options

Setting	Description	Default
Entity Accuracy	Confidence threshold (0-100%)	80%
Number of Pages	SERP pages to analyze	1
Top Entities	Entities to display in chart	15

Output

CSV file with columns:

• entity - The extracted entity
• confidence - Confidence score (0-1)
• category - Entity type/category
• wiki_url - Wikipedia link
• most_frequent / # of mentions - Frequency count

Use Cases

• Content Optimization: Find entities competitors mention
• Keyword Research: Discover related topics
• Content Gaps: Find entities you're not covering
• Video Research: Extract topics from YouTube content

Author

Lee Foot - eCommerce SEO Consultant