Understand User's Intent from Speech and Text

http://research.microsoft.com/en-us/projects/IntentUnderstanding/

Understanding what users like to do/need to get is critical in human computer interaction. When natural user interface like speech or natural language is used in human-computer interaction, such as in a spoken dialogue system or with an internet search engine, language understanding becomes an important issue.  Intent understanding is about indentifying the action a user wants a computer to take or the information she/he would like to obtain, conveyed in a spoken utterance or a text query.

 

In this project, we develop robust data-driven technologies applicable todifferent domains, make them morepractical by leveraging large amount of unlabeled data via unsupervised/semi-supervised machine learning;by innovating machine learning algorithms that work better with less data or mismatched data; and by augmenting statistical models with domain knowledge obtainedin a semi-supervised fashion.Research activities fall into the following areas:

• Data-Driven Approaches to Spoken Language/Query Understanding
• Unsupervised/Semi-Supervised Learning
• Automatic/Semi-automatic Acquisition of Domain Knowledge
• Authoring Tools for Spoken Language Understanding
• Application of Intent Undrestanding Technology

We have contributed to Microsoft products from the following teams:

• Microsoft Live Search/Commerce Search
• Microsoft adCenter
• Microsoft Speech Component Group
• Tellme

 

People

Li Deng    

Y. C. Ju    

Dong Yu    

Geoffrey Zweig    

 

Publications

• Larry Heck, Dilek Hakkani-Tur, Madhu Chinthakunta, Gokhan Tur, Rukmini Iyer, Partha Parthasarathy, Lisa Stifelman, Elizabeth Shriberg, and  Ashley Fidler, Multimodal Conversational Search and Browse, IEEE Workshop on Speech, Language and Audio in Multimedia,  August 2013
• Malcolm Slaney, Pay Attention, Please: Attention at the Telluride Neuromorphic Cognition Workshop, in IEEE SLTC Newsletter, IEEE,  November 2012
• Jingjing Liu, Xiao Li, Alex Acero, and  Ye-Yi Wang, Lexicon Modeling for Query Understanding, in ICASSP, IEEE,  May 2011
• Xiao Li, Understanding the Semantic Structure of Noun Phrase Queries, in ACL, Association for Computational Linguistics,  July 2010
• Xiao Li, Ye-Yi Wang, Dou Shen, and  Alex Acero, Learning with Click Graph for Query Intent Classification, in ACM Transaction on Information Systems, vol. 28, no. 3, Association for Computing Machinery, Inc.,  June 2010
• Mehdi Hafezi Manshadi and  Xiao Li, Semantic Tagging of Web Search Queries, in ACL, Association for Computational Linguistics,  August 2009
• Xiao Li, On the Use of Virtual Evidence in Conditional Random Fields, in EMNLP,  August 2009
• Xiao Li, Ye-Yi Wang, and  Alex Acero, Extracting Structured Information from User Queries with Semi-Supervised Conditional Random Fields, in SIGIR,  July 2009
• Young-In Song, Ye-Yi Wang, Yun-Cheng Ju, Mike Seltzer, Ivan Tashev, and  Alex Acero, Voice Search of Structured Media Data, in International Conference on Acoustics, Speech and Signal Processing, Institute of Electrical and Electornic Engineers, Inc., Taipei, Taiwan,  April 2009
• Ye-Yi Wang, Xiao Li, and  Alex Acero, Inductive and Example-Based Learning for Text Classification, in Interspeech, International Speech Communication Association, Brisbane, Australia,  September 2008