Sub-Model Generation to Improve Classification Accuracy
作者： Rajesh Balchandran , Gregory Purdy , Linda M. Boyer
DOI:
关键词: Linear classifier 、 Statistical classification 、 Multiclass classification 、 One-class classification 、 Pattern recognition 、 Computer science 、 Artificial intelligence 、 Classification rule 、 Statistical model 、 Natural language understanding
摘要: A method of classifying text input for use with a natural language understanding system can include determining classification information including primary and one or more secondary classifications received using statistical model (statistical model). sub-model sub-model) be selectively built according to generation criterion applied the information. The further selecting as final outputting input.
lens.org 参考文章(31)
Mark Epstein, Hierarchical language models ,(2001)
Chang-Ning Huang, Jianfeng Gao, Mu Li, Ming Zhou, Jian Sun, Lei Zhang, Method and apparatus using source-channel models for word segmentation ,(2003)
Gareth Loudon, Haizhou Li, Horng Jyh Paul Wu, Shuanhu Bai, System for chinese tokenization and named entity recognition ,(1999)
Kishor Morkhandikar, Pallaki Gururaj, Bandi Ramesh Babu, Ian M. Bennett, Intelligent query engine for processing voice based queries ,(1999)
Mazin G. Rahim, Srinivas Bangalore, Narendra K. Gupta, System and method of spoken language understanding in human computer dialogs ,(2002)
William D. Ramsey, Jonas Barklund, Sanjeev Katariya, Adaptive task framework ,(2005)
Rajesh Balchandran, Gregory Purdy, Linda M. Boyer, Information Extraction in a Natural Language Understanding System ,(2007)
Frederic Bechet, Dilek Hakkani-Tur, Jeremy Wright, Allen Gorin, Method and system for creating a named entity language model ,(2003)
Domenic Cipollone, YeYi Wang, Alejandro Acero, Leon Wong, Ciprian Chelba, Michael Calcagno, Ravi Shahani, Curtis Huttenhower, Statistical classifiers for spoken language understanding and command/control scenarios ,(2003)
Eric William Brown, Anni Rosa Coden, System, method and apparatus providing collateral information for a video/audio stream ,(2000)