报告题目：New Advances in Context Aware Natural Language Processing 报告人：Prof. Joyce Chai, Michigan State University 摘要： The Language and Interaction Research (LAIR) Group at Michigan State University explores new opportunities and research advances to incorporate context of language use in automated language understanding. The group is currently investigating linguistic discourse for semantic analysis, conversation context for automated inference about conversation participants, and non-verbal modalities for situated language processing. In this talk, I will focus on two of our on-going projects. The first is on automated identification of implicit arguments for nominal semantic roles. Our studies have shown that, compared to verbs, nominal predicates tend to have more implicit arguments - arguments that are not explicitly specified within a sentence. These implicit arguments often serve critical roles for downstream semantic analysis. How to automatically identify these implicit arguments becomes an important yet challenging task. I will discuss our initial results on this new emerging problem. The second part of my talk addresses the incorporation of visual context in situated language processing. Motivated by psycholinguistic findings that human eye gaze tightly links to human language production, we are developing approaches to incorporate perceived visual context into different levels of language processing. I will talk about these approaches, particularly to reference resolution and vocabulary acquisition, and discuss our empirical findings and results. 简历:
|
| 今天是： |
Joyce Chai is an Associate Professor in the Department of Computer Science and Engineering at Michigan State University. She received Ph.D. in Computer Science from Duke University in 1998. Prior to joining MSU in 2003, she was a Research Staff Member at IBM T. J. Watson Research Center. Her research interests include natural language processing, multimodal conversational systems, and intelligent user interfaces. She is a receipient of National Science Foundation Career Award in 2004 and ACL Best Long Paper Award in 2010.