Towards Context-adaptive Natural Language Processing Systems
作者： Michael Strube , Robert Porzel
DOI:
关键词:
摘要: In this paper we will discuss several issues and requirements for enabling natural language processing systems to become context-adaptive. Given the fact that emerging feature speaker independent continuous speech recognition restricted individual domains are equipped with syntactic semantic parsers understanding input from these domains, should begin envision open multi-domain systems. We address following arising in such systems: How divide modules context-variant context-invariant components how enable ensuing detect contextual changes phenomena.
参考文章(24)
Cecile L. Paris, User Modeling in Text Generation St. Martin's Press, Inc.. ,(1993)
Rainer Malaka, Robert Porzel, Integration of Smart Components for Building Your Personal Mobile Guide ,(2002)
Alex Waibel, Martin Westphal, Towards spontaneous speech recognition for on-board car navigation and information systems. conference of the international speech communication association. ,(1999)
Heike M. Buhl, Wissenserwerb und Raumreferenz: Ein sprachpsychologischer Zugang zur mentalen Repräsentation DE GRUYTER. ,(1996) , 10.1515/9783110929676
Alex Waibel, Tanja Schultz, Language adaptive LVCSR through Polyphone Decision Tree Specialization Workshop on Multi-lingual Interoperability in Speech Technology. pp. 85- 90 ,(2000)
Christian Kray, Anselm Blocher, Modeling the Basic Meanings of Path Relations international joint conference on artificial intelligence. pp. 384- 393 ,(1999)
Bob Carpenter, The logic of typed feature structures Cambridge Tracts in Theoretical Computer Science 32. ,(1992)
Mary Carroll, Deictic and Intrinsic Orientation in Spatial Descriptions: A Comparison Between English and German Advances in Psychology. ,vol. 103, pp. 23- 44 ,(1993) , 10.1016/S0166-4115(08)61655-6
Rainer Malaka, Alexander Zipf, DEEP MAP: Challenging IT Research In The Framework Of A Tourist Information System Springer, Vienna. pp. 15- 27 ,(2000) , 10.1007/978-3-7091-6291-0_2
Jerome A. Feldman, Srinivas Sankara Narayanan, Karma: knowledge-based active representations for metaphor and aspect University of California, Berkeley. ,(1997)