Weiwei Sun – A Computational Linguist

I am a Computational Linguist. I am a senior lecturer (=associate professor in US) at the Department of Computer Science and Technology of Cambridge University. Before coming to Cambridge, I were an associate professor at Wangxuan Institute of Computer Technology of Peking University. I completed my PHD at Saarland University under the supervision of Prof. Hans Uszkoreit. I obtained my BA in Linguistics, and my BS and MS in Computer Science from Peking University.

Here is my CV.

Research

The scientific study of language requires a blend of quantitative and symbolic computational approaches, combined with theoretical interest and expertise. Theoretically, the exploration of quality computational models of human language gives insights into the explanation of linguistic phenomena. Empirically, well-implemented computational models play an essential role in testing the validity of linguistic as well as psycholinguistic hypotheses. Furthermore, fruits of computational inquiry of human language benefit language technology and help create highly reliable automatic systems that master some aspects of human language use. Motivated by the above functions, my research is strongly interdisciplinary, leveraging computational models to investigate fundamental linguistic questions and developing linguistically-motivated technologies to advance Natural Language Processing applications.

I believe that the scientific and engineering goals of computational linguistics share a common ground — modeling the complexity of natural languages in sophisticated linguistic representations precisely. My research concentrates on applying graphs to encode syntactic and semantic analyses and exploiting graph-centric formalisms and algorithms to formulate linguistic theories explicitly. In order to build a mathematical model for linguistic analysis, we need precise formalisms and sound algorithms to manipulate graphs in a principled way. To extend the work to an empirical domain, we need comprehensive resource grammars and additional high-quality corpora. I study all of these topics. I am an active member of DELPH-IN.

My interest in human language has also led me to develop innovative technologies for intelligent computer-aided language learning (iCALL). I believe that high-performance natural language understanding systems for learner languages facilitate iCALL applications, such as grammatical error correction and automatic essay scoring. However, almost all systems are developed for first languages and perform rather poorly on learner texts. More recently, I have been working on syntactic and semantic processing for learner languages.

Teaching

I taught a tutorial on Graph-Based Meaning Representations: Design and Processing with Alexander Koller and Stephan Oepen at ACL 2019.

I teach L90: Overview of Natural Language Processing at Cambridge.

I taught six courses at Peking University:

  • Empirical Methods in Natural Language Processing

  • Programming and Linguistic Analysis

  • Compilers

  • Language, Logic and Computation

  • Introduction to Formal Syntax

  • Syntactic Theories and Grammar Engineering