Hitched Wedding Vendor Data Extraction and AI-Powered Name Extraction

Key Features

• Scrapes vendor data from multiple pages on Hitched.co.uk.
• Extracts first and last names from vendor descriptions using Azure OpenAI.
• Generates JSONL and CSV outputs for easy data handling.
• Supports concurrent web scraping for fast data collection.

Technologies Used

Challenges

Ensuring accuracy in AI-powered name extraction from unstructured text.Handling large datasets efficiently using multi-threaded scraping.Integrating AI output with traditional web-scraped data formats (CSV, JSONL).

Solution

Leveraged **Azure OpenAI via LangChain** to extract names from vendor descriptions, integrated multi-threading with **ThreadPoolExecutor** to speed up data scraping, and used **Pydantic** to ensure structured data output in both **JSONL** and **CSV** formats.

Results

The project efficiently collected and augmented vendor data from Hitched.co.uk, enabling automated generation of enriched vendor profiles ready for further analysis or integration into marketing systems.