Michael Gubanov

E-mail: gubanov.mike at gmail.com
Phone: +1-650-2151312

About

 I am starting my postdoc. I have earned my Ph.D. in Computer Science from the University of Washington (Seattle, WA) in collaboration with IBM Almaden Research Center (Almaden, CA) and Stanford University (Stanford, CA). I also spent some time at Google (Mountain View, CA) doing research on Big data, Machine learning, Data integration, and new Biomedical applications.

My current research is in Big data, Cloud computing, and Bioinformatics, with contributions to large scale Data Management, Information Retrieval, Machine Learning, and Natural Language Processing. Before coming to Seattle I earned my B.Sc. from St. Petersburg University of IT, Mechanics, and Optics (ACM World champions).

Book chapters (invited)
ReadFast: Structural Information Retrieval from Biomedical Big Text by Natural Language Processing [bib]
Michael Gubanov, Linda Shapiro, Anna Payt
Invited book chapter in "Information Reuse And Integration In Academia And Industry", Springer, 2013
Simplifying Information Integration: Object-based flow-of-mappings framework for integration [bib]
Bogdan Alexe, Michael Gubanov, Mauricio A. Hernandez, Howard Ho, Jen-Wei Huang, Yannis Katsis, Lucian Popa, Barna Saha, Ioana Stanoi
Invited book chapter in "Business Intelligence for the Real Time Enterprise", Springer, 2009

Selected publications (full list)

    BigDB: Automatic machine learning optimizer [pdf]
    Michael Gubanov, Anna Pyat
    arXiv:1301.1575 [cs.DB], 2013

     MedReadFast: Structural Information Retrieval Engine for Big Clinical Text [bib] [pdf]
    Michael Gubanov, Anna Pyat
    Proceedings of the 13th IEEE International Conference on Information Reuse and Integration (IRI), Las Vegas, Nevada, 2012, acc. rate 27%

     Using Unified Famous Objects (UFO) to Automate Alzheimer's Disease Diagnostics. [bib] [pdf]
    Michael Gubanov, Linda Shapiro
    Proceedings of the IEEE International Conference on Bioinformatics & Biomedicine (BIBM), Atlanta, Georgia, 2011

     ReadFast: Browsing large documents through Unified Famous Objects (UFO). [bib] [pdf]
    Michael Gubanov, Linda Shapiro, Anna Payt
    Proceedings of the 12th IEEE International Conference on Information Reuse and Integration (IRI), Las Vegas, Nevada, 2011, acc. rate 29%

    Learning Unified Famous Objects (UFO) to Bootstrap Information Integration. [bib] [pdf] [book]
    Michael Gubanov, Linda Shapiro, Anna Payt.
    Proceedings of the 12th IEEE International Conference on Information Reuse and Integration (IRI), Las Vegas, Nevada, 2011, acc. rate 29%

    IBM UFO Repository: Object-oriented data integration [bib] [pdf] [book]
    Michael Gubanov, Lucian Popa, Howard Ho, Hamid Pirahesh, Jeng-Yih Chang, Shr-Chang Chen.
    Proceedings of the 35th International Conference on Very Large Data Bases (VLDB), Lyon, France 2009, acc. rate 27%

    Metadata Management Engine for Data Integration with Reverse-Engineering Support [bib][pdf]
    Michael Gubanov, Phil Berstein, Alex Moshchuk
    Proceedings of the 2008 IEEE 24th International Conference on Data Engineering (ICDE), Cancun, Mexico 2008, acc. rate 29%

Selected research projects
ReadFast - a new structural Information Retrieval (IR) engine for Big text. Automatically extracts text schema using NLP that is used to retrieve information in efficient manner. Invited for publication as a book chapter by Springer in 2012.
Unified Famous Objects (UFO) - a new self-learning data integration technology.
Automatically locates and fuses needed data fragments from distributed data sources. Highlighted in the book on data management published by Springer in 2011.
RankHive During my internship at Google I designed and implemented components of a Big machine learning system for Web data. Used in many Google products inluding product search, Web crawler, machine translation, etc.

Talks (invited)
    Towards gaining control over Big medical data.
    University of California, Irvine, CA, 2012.

    Towards gaining control over information overflow.
    University of Central Florida, Orlando, FL, 2012.

    Towards gaining control over information overflow.
    University of Kentucky, Lexington, KY, 2012.

    Simplifying access to structured and unstructured data.
    Stanford University, Stanford, CA, 2011.

    Object-oriented management of structured and unstructured data.
    University of Washington, Seattle, WA, 2010.

    Simplifying information integration using Unified Famous Objects (UFO).
    University of Washington, Seattle, WA, 2010.

    IBM UFO Repository.
    IBM Almaden Research Center, San Jose, CA, 2007.

    Improving local search ranking with NLP.
    Mcrosoft, Seattle, WA, 2006.

    Feature engineering architecture for Web-search ranking.
    Mcrosoft, Seattle, WA, 2006.

    Improving ranking of product search results.
    Google, Mountain View, CA, 2005.

Selected awards
Best poster and travel award: "Using Online Machine Learning and Stream Processing to Detect Blood Coagulation"
 
Travel award: "Using Online Machine Learning and Stream Processing to Detect Hemoglobin in Plasma"
 
Clarendon Fund Full Felowship for 3 years, Oxford University, UK
George Soros national award for research excellence twice in a row
Winner of Russian national physics contest

Links
© Copyright 2011 Michael Gubanov. All rights reserved.