|
|
Xiaoxiao Shi |
|
Department of Computer
Science |
|
|
Office: 851 S. Morgan St., Rm 1336 SEO, Chicago, IL 60607 |
|
|
Office Phone:
(+1) 312-413-1270
Email:
xiao.x.shi@gmail.com or
xshi9@uic.edu |
Biography
2009.8~present Department of Computer Science, University of Illinois at Chicago, with supervisor Prof. Philip S. Yu.
2010.8~present Applied Math (Concurrent Degree) at Dept. of Mathematics, Statistics, and Computer Science, University of Illinois at Chicago.
2008.9~2009.2 Visiting student in Department of Computer Science, HKUST, with supervisor Prof. Qiang Yang.
2007.9~2009.6 Master in the Department of Computer Science, Sun Yat-sen University (Zhongshan University).
2003.9~2007.7 B.Eng in the Department of Computer Science, Sun Yat-sen University (Zhongshan University).
Research Interests
Statistical Cascading Relationship Inference:
Learning from Heterogeneous Sources:
Others: Social Network Mining,
Bioinformatics, Privacy Preserving Data
Publishing, Active Transfer Learning, Active Learning.
It aims at discovering causality relationships given a set of evolving objects (time-series data). For example, given the historical stock prices, how can we discover who affects whom? What are the most influential stocks?
(a) Learning from data with different statistical properties: Data Stream
Mining, Sample Selection Bias, Multi-task
Learning, Transfer Learning, Heterogeneous Learning.
(b) Learning from data with different structures: Collective Classification,
Graph Mining, Feature Generation.
Selected
Publication
[2012] Guan Wang, Yuchen Zhao, Xiaoxiao Shi, and Philip S. Yu, "Magnent Community Identification on Social Networks", to appear in KDD'2012.
[2012] Xiaoxiao Shi, Jean-Francois Paiement, David Grangier, and
Philip S. Yu, "Learning from Heterogeneous Sources via Gradient Boosting Consensus", to appear in SDM'2012 (regular paper). The processed heterogeneous dataset can be found here.
[2012] Xiaoxiao Shi, Xiangnan Kong, and
Philip S. Yu, "Transfer Significant Subgraphs across Graph Databases", to appear in SDM'2012 (regular paper).
[2011] Xiaoxiao Shi, Qi Liu, Wei Fan, and
Philip S. Yu, "Transfer across Completely Different Feature Spaces via Spectral Embedding", accepted by Transactions on Knowledge and Data Engineering (TKDE). The processed dataset can be found here.
[2011] Xiaoxiao Shi, Yao Li, and
Philip S. Yu, "Collective Prediction with Latent Graphs", ACM Conference on Information and Knowledge Management (CIKM'11), lasgow, Scotland, UK, 2011.
[2011] Xiaoxiao Shi, Wei Fan, Jianping Zhang, and
Philip S. Yu, "Discovering Shaker from Evolving Entities via Cascading Graph Inference", 17th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD'11), San Diego, CA, 2011. The processed dataset can be found here.
[2011] Xiaoxiao Shi and
Philip S. Yu, "Limitations of Matrix Completion via Trace Norm Minimization", SIGKDD Explorations, 12(2):16-20, 2011.
[2011] Xiangnan Kong, Xiaoxiao Shi, and
Philip S. Yu, "Multi-label Collective Classification", 2011 SIAM International Conference on Data Mining (SDM'11), Mesa AZ, 2011.
[2010] Xiaoxiao Shi, Wei Fan, and
Philip S. Yu, "Efficient Semi-supervised Spectral Co-clustering with
Constraints", 2010 IEEE International Conference on Data Mining (ICDM'10), 2010. The
Powerpoint presentation can be found here.
[2010] Xiaoxiao Shi, Qi Liu, Wei Fan, Philip S. Yu, and Ruixin Zhu,
"Transfer Learning on Heterogenous Feature Spaces via Spectral Transformation",
2010 IEEE International Conference on Data Mining (ICDM'10), 2010. The
Powerpoint presentation can be found here.
[2010] Xiaoxiao Shi, Kevin Chang,
Vijay K. Narayanan, Vanja Josifovski and Alex J. Smola, "A Compression Framework
for Generating User Profiles", 2010 ACM SIGIR workshop on feature generation and
selection for information retrieval, Geneva, Switzerland, July, 2010.
[2010] Xiaoxiao Shi, Qi Liu, Wei
Fan, Qiang Yang and Philip S. Yu, "Predictive Modeling
with Heterogeneous Sources", 2010 SIAM International Conference
on Data Mining (SDM'10), Columbus, Ohio, Apr 2010. The
Powerpoint presentation is here. The
clustering-based KL divergence code can be found at
here.
[2009] Xiaoxiao Shi, Wei Fan, Qiang Yang
and Jiangtao Ren, "Relaxed Transfer of Different
Classes via Spectral Partition" , 2009 European Conference on
Machine Learning and Principles and Practice of Knowledge Discovery in Databases
(ECML/PKDD 2009) September 7-11, 2009, Bled, Slovenia.
The Powerpoint presentation is
here. The
relaxed transfer code &
the clustering-based KL divergence code can be
found at here.
[2008] Xiaoxiao Shi,
Wei Fan, and Jiangtao Ren, "Actively
Transfer Domain Knowledge", 2008 European Confernce on Machine Learning and
Principles and Practices of Knowledge Discovery in Databases (ECML/PKDD08),
Antwerp, Belgium. The Powerpoint presentation is
here. The
code and dataset (synthetic and landmine) written by Xiaoxiao Shi can be found
at here, and the 20 Newsgroup data can be found at
here. (Active Transfer Learning)
[2008] Jiangtao Ren, Xiaoxiao Shi,
Wei Fan, and Philip S. Yu "Type
Independent Correction of Sample Selection Bias via Structural Discovery and
Re-balancing", 2008 SIAM International Conference on Data Mining
(SDM'08), Atlanta, GA, Apr 2008.
Honors & Awards
2nd Place of the intern's hackday, hosted by LinkedIn (170 hackers), 2011 (Winning project: LinkedOut).
6th Place of ACM ICPC Programming Contest. Guangdong Province, 2006.
Hewlett Packard Fellowship. University of Illinois at Chicago, 2011 to 2012.
Wexler Award. University of Illinois at Chicago, 2010 to 2011.
University Fellowship. University of Illinois at Chicago, 2009 to 2010.
Departmental Fellowship. Department of Computer Science, University of Illinois at Chicago, 2009 to 2010.
Student Travel Award, SDM'10, KDD'11.
First Scholarship (1st place), Sun Yat-sen University, 2003 to 2007.
Excellent Undergraduate Student, Sun Yat-sen University, 2007.
Excellent Graduate Student, Sun Yat-sen University, 2009.
Teaching Assistant
MSC-IT 521, Fall 2008: Knowledge Discovery and Data Mining at HKUST
CS 408, Fall 2010: Database Systems at UIC
CS 411, Fall 2010: Artificial Intelligence at UIC
