Abstract
This chapter describes the applications and natural language corpora used to develop and to validate the semantic decoding strategies explored in this work. The applications concern air- and train-travel information retrieval and reservation as well as an appointment scheduling task. The work makes use of periodic evaluations to monitor progress through objective performance measures. Current criteria for spoken language systems evaluation are also presented.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
AREA is the central research organization for the United States Department of Defense. It pursues innovative research and development projects having significant potential for both military and commercial applications and supports a national technology base that serves both civilian and military purposes. The technological impact of the research goes beyond the normal evolutionary developmental approaches; the ARPA mission is to pursue these ideas from the demonstration of technical feasibility through the development of prototype systems.
The research was carried out in collaboration with the MIT-LCS Spoken Language Systems Group.
NLPARSE is a Texas Instruments propriety system, made available to the ARPA research community for the ATIS application. In a WOZ setup the wizard input, an NL-parse paraphrase, is provided to NLPARSE which then simulates a system response to the user.
Train à Grande Vitesse (High Speed Train)
The weighted response error is calculated as the sum of #(no_answer) and 2 * #(wrong_answer)
In the following, the closest translation possible for the French example sentences is provided in parentheses.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer Science+Business Media New York
About this chapter
Cite this chapter
Minker, W., Waibel, A., Mariani, J. (1999). Applications and Corpora. In: Stochastically-Based Semantic Analysis. The Springer International Series in Engineering and Computer Science, vol 514. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-5255-0_3
Download citation
DOI: https://doi.org/10.1007/978-1-4615-5255-0_3
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-7396-4
Online ISBN: 978-1-4615-5255-0
eBook Packages: Springer Book Archive