Sui, Zhifang is a professor in the School of Computer Science. She obtained his B.Sc. from Shandong University in 1992, and Ph.D. from Peking University in 1998 respectively. Her research interests include Natural Language Processing, Text Mining and Language Knowledge Engineering.
Dr. Sui has published more than 60 research papers, and most of them are published in top-tier conferences and journals, such as ACL, IJCAI, EMNLP and COLING. She has served as PC Co-Chair of CLSW 2013 and is a Council Member of Chinese Information Processing Society of China. She was awarded the second prize of National Scientific and Technological Progress Award in 2011.
Dr. Sui has about ten research projects including NSFC, 973 programs, and the National Social Science Foundation in China etc. Her research achievements are summarized as follows:
1) Language Knowledge Engineering:One of the critical problems in knowledge engineering is that, there is gap between the shallow structures expressed by natural languages and the deep structures in conceptual knowledge By using binary conceptual relations and their language expressions as the research focus, Dr. Sui proposes to using qualia structure and thematic structure to respectively model the conceptual semantics of nouns and verbs, we aims to conduct systematic research on establishing the mapping resources between conceptual relations and their language expressions under the guidance of the theory of linguistics and cognitive linguistics.
2) Text Mining for Domain Knowledge Engineering: As an important tasks in ontology learning, instance extraction and attribute extraction of the concepts have attracted more research attentions. Dr.Sui puts forward a weakly-supervised method which can synchronously extract instances and attributes for a concept based on web information. There are close relations between the instances and the attributes of a concept. To detect an instance will be beneficial for recognizing the attributes. On the other hand, to recognize the attributes will also help to extract the instances for a concept.
3) Multi-level Deep Semantic Parsing of Chinese Sentences: Deep semantic parsing of Chinese sentences aims to understand sentences more thoroughly and deeper. To promote the research on Chinese deep semantic parsing, Dr. Sui proposes a multi-level deep semantic parsing of Chinese sentences. This task is unique in the following aspects: (1) it focuses on Chinese, aiming to build a paradigm for describing the semantics of Chinese sentences by considering the characteristics of Chinese sentences; (2) Semantic interpretation of natural language usually involves multiple levels including proposition semantic level, complemental logic semantic level, entity relation level, event relation level, inter-sentence relation level and construction semantic level. This task reveals the deep semantics of Chinese sentences progressively through layer-wise annotations.
