Zheng (Alan) Zhao

"You hear and you forget; you see and you remember; you do and you understand." - Confucius



Zheng Zhao Ph.D

1600 Amphitheater Pkwy, Mountain View, CA, 94043 

E-mail: alzhao (AT) google (DOT) com or alanzhao (AT) gmail (DOT) com

ABOUT ME

I am a software engineer at Google Inc. My work is to develop large scale machine learning models to help Google serve more relevant display ads. I was a senior research statistician at the SAS Institute Inc from 2010 to 2015I developed the SAS High Performance Analytical Procedures: PROC HPREDUCE, PROC HPTMINEPROC HPTMSCORE, and PROC HPBOOLRULEI received my Ph.D. in Computer Science and Engineering from Arizona State University (ASU) in 2010, and my M.Eng. and B.Eng. from Harbin Institute of Technology (HIT), P. R. China in 2002 and 2000, respectively. 


RESEARCH INTERESTS

High-performance analytics methods for handling very large-scale structured/unstructured data of very high dimensionality.

  • Click Through Rate and Conversion Rate Prediction
  • High Performance Data Mining & Text Mining
  • Dimensionality Reduction
  • Deep Learning
      Spectral Feature Selection for Data Mining


      BOOK
      Zheng (Alan) Zhao and Huan Liu. Spectral Feature Selection for Data Mining, Chapman and Hall/CRC, 2011.

      You can obtain the book from CRC Press (EZM02 for 25% off) and Amazon.
      Please feel free to contact us if you have found any "good", "bad" or "ugly" in the book.


      Recent Publications (Complete list can be found here)
      • Zheng Zhao, Jun Liu, James Cox. Safe and Efficient Screening for Sparse Support Vector Machine. In: Proceedings of the 20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2014.
      • Zheng Zhao, Jun Liu, James Cox. Accelerating Model Selection with Safe Screening for L1-Regularized L2-SVM. In: Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD), 2014.
      • Jun Liu, Zheng Zhao, Jie Wang, and Jieping Ye. Safe Screening with Variational Inequalities and Its Application to Lasso. The 31st International Conference on Machine Learning (ICML), 2014
      • Zheng Zhao, Russell Albright, and James Cox. Processing and Storing Sparse Data in SAS® Using SAS® Text Miner Procedures, SAS Global Forum, 2014.
      • Zheng Zhao, Ruiwen Zhang, James Cox, David Duling and Warren Sarle. Massively Parallel Feature Selection: An Approach Based on Variance Preservation, Machine Learning, 92(1), 195-220, 2013.
      • Zheng Zhao, Russell Albright, James Cox and Alicia Bieringer. Big Data Meets Text Mining, SAS Global Forum, 2013.


      Highly Cited Works (a more comprehensive list can be found at Google Scholar)

      Software, Source Code and Data Sets
      • SAS High Performance Analytical Procedures: PROC HPREDUCEPROC HPTMINE, and PROC HPTMSCORE. (link)
      • Feature Selection Repository at ASU (link)
      • INTERACT for finding interacting features (link)
      • MRSF, Efficient Spectral Feature Selection with Minimum Redundancy (link)
      • SPEC, Univariate Spectral Feature Selection (link)