Speaker: Shanfeng Zhu

PhD Associate professor, School of Computer Science,Fudan University.
Email: zhusf@fudan.edu.cn
Web：http://datamining-iip.fudan.edu.cn/
Background:
1999-2003 Ph.D. Degree, Department of Computer Science, City University of Hong Kong, P.R, China
1996-1999 M.Sc. Degree, Department of Computer Science, Wuhan University, P.R, China
1992-1996 B.Sc. Degree, Department of Computer Science, Wuhan University, P.R, China

Research:

Data Mining, Machine Learning, Information Retrieval, Intelligent Information Processing
简介：
朱山风，复旦大学计算机科学技术学院副教授，博士生生导师。香港城市大学博士（2003）,日本京都大学博士后（2004-2008）,日本学术振兴会邀请访问学者（JSPS Invitation Fellowship 2012）,美国伊利诺伊 大学香槟分校访问学者（2013-2014）,日本京都大学访问副教授（2016）。主要研究方向为生物信息学、信息检索和数据挖掘。在相关领域的著名国际期刊和会议如KDD、IJCAI、ISMB、Bioinformatics、NAR、Briefings in Bioinformatics等以第一作者或通讯作者发表论文50余篇。主持两项国家自然科学基金面上项目：大规模生物医学文献医学主题词的高精度自动标注研究（61572139）和MHC II类分子亲和肽的高精度预测研究 （61170097已结题）；一项国家自然科学基金青年项目：基于信息融合的生物医学文本高性能聚类研究（60903076已结题）。BIBM2014-2017、InCoB2012-2017、GIW2015-2017、APBC2014-2018等生物信息学国际会议程序委 员会委员。中国计算机学会计算机术语审定工作委员会委员（2010-2015），负责生物信息学名词审定。中国人工智能学会生物信息与人工生命专业委员会初始委员、中国计算机学会生物信息专业委员会初始委员，中国中 文信息处理学会医疗健康与生物信息处理专业委员会初始委员，中国运筹学会计算系统生物学分会理事。
Time : 2:30-4:30 pm , Jan. 12（Friday）
Venue: Room 300, SIBS Main Building, Yueyang Road 320
Host： Prof. Sijia Wang

CAS-MPG Partner Institute for Computational Biology
Title：MeSHLabeler and GOLabeler, Recent Progress in Large-Scale MeSH Indexing and Protein Function Prediction
Abstract:

Many important problems in BioCuration can be modeled as a large scale multi-label learning problem, such as MeSH indexing and protein function prediction. By utilizing learning to rank framework, we have developed MeSHLabeler and DeepMeSH to solve large-scale MeSH indexing problem, and GOLabeler for protein function prediction. DeepMeSH achieved the first place in both BioASQ4 and BioASQ5 challenge, and MeSHLabeler achieved the first place in both BioASQ2 and BioASQ3 challenges. Specifically, DeepMeSH achieved a Micro F-measure of 0.6323, 2% higher than 0.6218 of MeSHLabeler and 12% higher than 0.5637 of MTI (NLM's official solution), for BioASQ3 challenge data with 6000 citations. on the other hand, the empirical results obtained by examining GOLabeler extensively and thoroughly by using large-scale datasets revealed numerous favorable aspects of GOLabeler, including significant performance advantage over state-of-the-art AFP (Automated Function Prediction) methods. According to the initial evaluation of CAFA3 (The Critical Assessment of protein Function Annotation algorithms) in July 2017, GOLabeler achieved the first place in terms of F-max out of around 200 submissions by around 50 labs all over the world.
All are welcome！