段俊文

讲师 硕士生导师

所在单位:计算机学院

学历:博士研究生毕业

性别:男

联系方式:jwduan@csu.edu.cn

学位:博士学位

在职信息:在职

毕业院校:哈尔滨工业大学

   

个人简介

段俊文,博士,现为中南大学计算机学院讲师、硕士生导师。博士毕业于哈尔滨工业大学计算机学院社会计算与信息检索研究中心,博士期间曾赴美国伊利诺伊大学香槟分校(UIUC)联合培养,导师ACM Fellow 翟成祥教授。主要研究内容为自然语言处理、文本挖掘、信息抽取、事理图谱、文本表示学习等。在国际知名期刊和会议TASLP/AAAI/IJCAI/EMNLP/COLING/NAACL等发表论文10余篇,同时担任ACL,AAAI, IJCAI, EMNLP等领域顶级会议和期刊审稿人,社会媒体处理专委会通讯委员。目前主持国家自然科学基金青年项目和湖南省重点实验室开放基金项目各一项,参与多个国家级项目的研究,包括国家重点基础研究发展计划 (973) 项目“中文深度分析与开放域内容理解”、科技创新2030—“新一代人工智能”重大项目“可泛化的领域知识学习与计算引擎”、国家自然科学基金重点项目“篇章级中文语义分析理论与方法”等。


【研究方向】围绕自然语言处理展开,具体包括:深度学习、文本表示学习、信息抽取(实体识别、关系抽取、事件抽取)、知识图谱与事理图谱的构造与应用、机器阅读理解、对话系统等。(什么是自然语言处理?)

硕士招生】每年招收3名左右硕士研究生,欢迎对自然语言处理、文本挖掘感兴趣的同学与我联系(邮箱:jwduan@csu.edu.cn)。也欢迎对自然语言处理感兴趣的优秀本科生加入课题组。


代表性论文


[1] Junwen Duan, Xiao Ding, Yue Zhang* and Ting Liu*. TEND: A Representation Learning Framework for News Documents. IEEE/ACM Transactins on Audio, Language and Speech, 2019. (CCF B 类期刊, JCR 一区)

[2] Junwen Duan, Xiao Ding, and Ting Liu*. A Guassian Copula Regression Model for Movie Box-office Revenues Prediction. Science China Information Science, 2016. (CCF B 类期刊)

[3] Junwen Duan, Yiheng Chen, Xiao Ding, and Ting Liu. Mining Intention-related Products on online QA Community, Journal of Computer Science and Technology, 2015. (CCF B 类期刊)

[4] Junwen Duan, Yue Zhang, Xiao Ding, Ching-Yun Chang, Ting Liu. Learning Targetspecific Representations of Financial News Documents for Cumulative Abnormal Return Prediction. Coling 2018. (CCF B 类会议)

[5] Junwen Duan, Xiao Ding, and Ting Liu. Learning Sentence Representations over Tree-structures for Target-Dependent Classification, NAACL 2018, pages 551-560. (CCF C 类会议)

[6] Junwen Duan, Xiao Ding, and Ting Liu. “A Gaussian Copula Regression Model for Movie Box-office Revenue Prediction with Social Media.” , In Social Media Processing, pages 28-37, Springer Singapore, 2015. (EI 检索)

[7] Junwen Duan, Xiao Ding, and Ting Liu. “Mining intention-related products on online q&a community.” , In Social Media Processing, pages 13-24. Springer, 2014 [Best Paper Honorable Mention] (EI 检索)

[8] Xiao Ding, Ting Liu, Junwen Duan, and Jian-Yun Nie. “Mining user consumption intention from social media using domain adaptive convolutional neural network., In Proceedings of Twenty-Ninth AAAI Conference on Artificial Intelligence, pages 2389–2395. AAAI Press, 2015. (CCF A 类会议

[9] Xiao Ding, Kuo Liao, Ting Liu, Zhongyang Li and Junwen Duan. Event Representation Learning Enhanced with External Commonsense Knowledge[C]//EMNLPIJCNLP. 2019. (CCF B 类会议)

[10] Xiao Ding, Yue Zhang, Ting Liu, and Junwen Duan. Knowledge-driven event embedding for stock prediction[C]//Proceedings of COLING 2016. (CCF B 类会议)

[11] Xiao Ding, Yue Zhang, Ting Liu, and Junwen Duan. “Deep learning for event driven stock prediction.” , In Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, pages 2327-2333. AAAI Press, 2015. (CCF A 类会议

[12] Xiao Ding, Yue Zhang, Ting Liu, and Junwen Duan. “Using structured events to predict stock price movement: An empirical investigation.” , In Proceedings of EMNLP 2014, pages 1415-1425. (CCF B 类会议