* Front page
  * Overview
     Requirements
  * Install
  * Screenshots
  * Documentation
     FAQ
     User guide
     Related links
     API
  * License
  * Download
  * Evaluation

Mustru: Documentation

FAQ

Q: What is Mustru ?
A: Mustru (Managing Unstructured Text & Retrieval Utility) is a desktop search engine . It uses Lucene (for the index) and Berkeley DB JE (for the database). The search interface is implemented using a JSP/Servlet.

Q: How does Mustru answer queries ?
A: Boolean queries are parsed and submitted to the underlying Lucene search engine. The top n results are returned in a hit list.

Q: How does Mustru answer questions ?
A: Questions are parsed into entities, transformation phrases, unigrams, and bigrams and submitted to Lucene. A list of passages are returned.

User Guide

The user guide is under development and should be available shortly.

Related Links

• WordNet: A dictionary / thesaurus
• Lingpipe: A Java API for NLP and computational linguistics
• GATE: A General Architecture for Text Engineering
• Berkeley DB JE: An embedded database with a Java API
• Lucene: Search engine API
• Antiword: Software to extract text from MS Word files
• POI: Java API to Access MS Office files
• JAVA: Technology to build and run the application
• OpenOffice: Office Suite for Linux and Windows