Team User's Guide
Abstract
TEAM (Transportable English Data Access Medium) is a transportable natural-language (NL) interface to a database. It is a tool of considerable power that enables the user to retrieve data and elicit answers to queries by asking questions and giving commands in English instead of a formal query language. Moreover, TEAM is not limited to any particular database, but can be adapted to demonstrate natural-language retrieval in a broad variety of application domains. The prototype TEAM software described herein was developed by the Artificial Intelligence Center of SRI International to demonstrate the system's capabilities and adaptive potential. This user's guide is designed to assist new TEAM users to learn about the concepts and tasks involved in retrieving data and in preparing a demonstration for a new application area. An effort has been made to illustrate some of the problems TEAM must solve in translating an English question into a database query. However, the necessarily limited scope of this guide cannot include a discussion of all the natural-language-processing issues addressed by the system; our emphasis is on a practical, rather than theoretical, understanding of the concepts. Similarly, while this guide cannot cover every detail of creating a new demonstration for TEAM, it does provide a thorough introduction to the procedure to be followed and explains how to use the on-line "help" provided by the system.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1984
- Accession Number
- ADA460927
Entities
People
- Lorna Shinkle
Organizations
- SRI International