Adequacy-fluency Evaluation of Natural Language

摘要:In this talk, we discuss the adequacy and fluency evaluation of natural language sentences. In particular, we study a two-dimensional automatic evaluation metric that is designed to operate at sentence level. The metric is based on the concepts of adequacy and fluency, aiming at evaluating semantic and syntactic content of a sentence individually and in combination. The evaluation metric provides a balanced view on the quality of a sentence regarding a reference. We will investigate the use of adequacy and fluency metric in machine translation evaluation, and dialogue system evaluation.

简历:Haizhou Li is a Professor at the Department of Electrical and Computer Engineering, National University of Singapore. His research interests include speech information processing, natural language processing, and neuromorphic computing. Professor Li has served as the Editor-in-Chief of IEEE/ACM Transactions on Audio, Speech and Language Processing (2015-2018), the President of the International Speech Communication Association (ISCA, 2015-2017), and the President of Asia Pacific Signal and Information Processing Association (APSIPA, 2015-2016). He is a Fellow of the IEEE and the ISCA. He was named Nokia Professor in 2009 by Nokia Foundation, and Bremen Excellence Chair Professor in 2019 by University of Bremen.