Welcome to Wayne Xin Zhao's Homepage!  

Brief Introduction

I'll graduate in July, 2014. I'm looking for a teaching position or a research position. Please contact me via batmanfly AT gmail.com if you had any positions.

Wayne Xin Zhao (赵鑫) is a Ph.D. candidate at CNDS Lab, Peking University (Beijing, China). 
Wayne, from the name of Batman(Bruce Wayne), is chosen as his English nick name.
Xin Zhao is a big fan of Batman.

He has a very broad range of research interests on various Web text mining tasks.

You can reach him by batmanfly AT gmail.com(EMAIL).

He has written a very short survey of topic models in text mining (in Chinese) Survey of topic models. and a note for LDA's variational methods vEM Note for LDA.


  • Ph.D. candidate: School of Electronics Engineering and Computer Science, Peking University, September 2008 - Now. His supervisor is Prof. Xiaoming Li.
  • Visiting student: School of Information Systems, Singapore Management University,March 2010 - Janunary 2011, where he was supervised by Assistant Prof. Jing Jiang.
  • Bachelor: Department of Computer Science and Technology, Harbin Institute of Technology, September 2004 - July 2008 (Ranking: 3rd of 183)
    • During the third year and fourth year, he had been studying in the IR-Lab of HIT.


Jinpeng Wang, Xin Zhao(**2nd) and Xiaoming Li.  Mining New Business Opportunities: Identifying Trend related Products by Leveraging Commercial Intents from Microblogs.  To appear in EMNLP'13, Long paper

Xudong Zhang, Xin Zhao(**2nd), Dongdong Shan, and Hongfei Yan.  Group-Scheme: A Universal SIMD-based Compression Scheme.  To appear in IEEE BigData'13, Short paper (6 pages)

Xin Zhao, Jinpeng Wang, Yulan He, Jian-Yun Nie and Xiaoming Li.  Originator or Propagator? Incorporating Social Role Theory into Topic Models for Twitter Content Analysis.  To appear in CIKM'13, Short paper (6 pages). [Local Copy][copyright belonging to ACM]  

Jinpeng Wang, Xin Zhao(*co-first authors), Yulan He and Xiaoming Li.  Infer User Interests via Link Structure Regularization.  To appear in ACM TIST

Xin Zhao, Yanwei Guo, Rui Yan, Yulan He and Xiaoming Li.  Timeline Generation with Social Attention.  To appear in SIGIR'13, Short paper

Yang Lu, Xin Zhao(**2nd), Hongfei Yan and Xiaoming Li.  A Metric Learning based approach for evaluating time series similarity.  To appear in WAIM'13, Full oral presentation

Yanwei Guo, Yuexin Wu, Xin Zhao (***third author) and Hongfei Yan.  A case study in online gaming service: the analysis and prediction of churn analysis and prediction.  In CCIR'13, Full Oral Presentation

Jinpeng Wang, Xin Zhao(**2nd), Rui Yan, Haitian Wei, Jianyun Nie and Xiaoming Li.  Using Lexical and Thematic Knowledge for Name Disambiguation.  In AIRS'12, Full oral presentation

Xin Zhao, Baihan Shu, Jing Jiang, Yang Song and Xiaoming Li.  Identifying Event-related Bursts via Social Media Activities.  In EMNLP'12, Full Oral Presentation

Yang Song, Jing Jiang, Xin Zhao(***3rd) and Houfeng Wang.  Joint Learning for Coreference Resolution with Markov Logic.  In EMNLP'12, Full Oral Presentation

Xin Zhao, Rishan Chen, Kai Fan, Hongfei Yan, Xiaoming Li.  A Novel Burst-based Text Representation Model for Scalable Event Detection.  To appear in ACL'12, Short Oral Presentation Our system

Dongdong Shan, Xin Zhao(**2nd), Rishan Chen, Baihan Shu, Hongfei Yan, Xiaoming Li.  EventSearch: A System for Event Discovery and Retrieval on Multi-Type Historical Data.  To appear in KDD'12, Demo. Our system 

Dongdong Shan, Xin Zhao(**2nd), Jing He, Hongfei Yan, Xiaoming Li.  Efficient Phrase Querying with Flat Position Index.    In CIKM'11, Poster

Jing He, Xin Zhao(**2nd), Baihan Shu, Xiaoming Li, Hongfei Yan.  Efficiently Collecting Relevance Information from Clickthroughs for Web Retrieval System Evaluation.  In SIGIR'11, Full Oral Presentation

Xin Zhao, Jing Jiang, Jing He, Yang Song, Palakorn Achanauparp, Ee-Peng Lim, Xiaoming Li.  Topical Keyphrase Extraction from Twitter.  In ACL'11, Full Oral Presentation

Xin Zhao, Jing Jiang, Jianshu Weng, Jing He, Ee-Peng Lim, Hongfei Yan, Xiaoming Li.  Comparing Twitter and Traditional Media using Topic Models.  In ECIR'11
Full Oral Presentation, Acceptance Rate = 20% .

Xin Zhao, Jing Jiang, Hongfei Yan, Xiaoming Li.  Jointly Modeling Aspects and Opinions with a MaxEnt-LDA Hybrid.  In EMNLP'10Full Oral Presentation, Acceptance Rate = 14%(Oral)+11%(Poster).
Xin Zhao, Jing Jiang, Jing He, Hongfei Yan, Xiaoming Li.  Context Modeling for Ranking and Tagging Bursty Features in Text Streams.  In CIKM'10, Short paperAcceptance Rate = 13.4%(full)+17.9%(short) .

Yu Zhang, Xin Zhao(** second author) and Ting Liu.  Research on Question Understanding for Cooperative Question Answering.  In Journal Of Chinese Information Processing,2008, (In Chinese)

Experiences and Professional Activities

WWW 2013, external reviewer.
CIKM 2012, external reviewer.
EMNLP'12 PC reviewer.
July 2011, SIGIR'11, Beijing, China.
June 2011, ACL'11, Portland, USA.
April 2011, ECIR'11, Dublin, Ireland.
March 25,2010 ~ Janunary 24, 2011. Visiting student in SMU. Supervisor: JIANG Jing, Singapore.
October 7, 2010 ~ October 12, 2010. EMNLP'10, Boston, USA.
November 22, 2010 ~ December 4, 2010. PhD leadership program held by Australia National University, Canberra, Australia.

Selected Awards

One of "Top ten student pivots in research of EECS, PKU, 2011".
Sohu Scholarship of PKU-NET Lab, 2011.
Google China PhD Fellowship, 2011 (one out of four).
The 1st place in the 2nd Heilongjiang Province programming contest, 2007.
Meritorious Winner of international Interdisciplinary Contest in Modeling, 2008. [results] (Team# 2457)
“The challenge cup” —Chinese college students business plan competition, silver medal in Hei Longjiang Contest District. 2008.
HP China Excellent Student Scholarship, 2006-2007.
Special Grade Scholarship of Harbin Insistute of Technology, 2005-2006.
Outstanding Graduate of Hei Longjiang Province, 2008.
Pivot of Merit Student of Harbin Institute of Technology, 2006.


  • 俞敏洪:"一个人的成长有两个条件,一是要给予足够的时间,没有任何人会在一天之内成长起来,就像一棵树一样,十年过去才会发现长大了,另一个条件是生命的热情永不熄灭,不管多么卑微,内心的种子一定要向往天空,要尽力伸展自己的枝叶去触摸蓝天,去追逐天空的云彩,有了向上的心,一定就会有灿烂的果。"
  • T.H. White:"治疗悲伤最有效的方法就是去学习一些东西。当你受到打击,经历挫折,从骨子里颤抖,看到你的世界被人摧毁,体验你的荣誉被踩进下水道,发生这些事情之后,你要学的只有一件事情。学习世事为何如此变化,发生了哪些变化。这是唯一可以让心永不疲惫,永不疏离,永不后悔的事情。"