Urdu Search Engine - NLP-Powered Search

Web (PHP + Lucene)

About

A crawler and web-based search engine for Urdu language websites, featuring unique NLP algorithms to distinguish between Persian, Arabic, and Urdu scripts with indexing and search result generation.

My Role

Led development of Urdu search engine with NLP algorithms for multi-script language processing at BZU Multan. Built crawler, indexer, and search result generation components.

Core Capabilities

• NLP algorithms for Urdu/Persian/Arabic language distinction
• Urdu language indexer with processing routines
• Web crawler for Urdu websites
• Search result generation from indexed data

Architecture

• PHP web application with MySQL backend
• Lucene.NET for search indexing
• Custom NLP processing pipeline

Tech Stack