Related Experiment Videos
A prototype natural language interface to a large complex knowledge base, the Foundational Model of Anatomy
Gregory Distelhorst1, Vishrut Srivastava, Cornelius Rosse
1Departments of Biological Structure and Medical Education and Biomedical Informatics, University of Washington, Seattle, WA 98195, USA.
AMIA ... Annual Symposium Proceedings. AMIA Symposium
|January 20, 2004
Summary
We developed GAPP, a natural language interface for querying the Foundational Model of Anatomy (FMA). This tool aids anatomy experts in evaluating the FMA knowledge base through structured queries.
Area of Science:
- Anatomy
- Computer Science
- Bioinformatics
Background:
- The Foundational Model of Anatomy (FMA) is a large knowledge base requiring efficient querying methods.
- Natural language interfaces can improve accessibility to complex anatomical databases.
Purpose of the Study:
- To develop and evaluate a constrained natural language interface for querying the FMA.
- To facilitate domain expert evaluation of the FMA knowledge base.
Main Methods:
- Developed GAPP, a natural language interface.
- Parsed user questions into subject-relation-object structures.
- Utilized domain-specific dictionaries to translate parsed sentences into StruQL queries.
- Implemented OQAFMA server to query the FMA and return results in XML.
Main Results:
- GAPP successfully handles simple and nested questions.
- The interface converts natural language queries into executable StruQL queries.
- Output is returned in XML format via the OQAFMA server.
Conclusions:
- GAPP demonstrates potential for enabling domain experts to evaluate the Foundational Model of Anatomy.
- The developed interface offers a novel approach to querying large anatomical knowledge bases.