index

 


me

  Yushi (Kevin) Jing(景宇石)   jing [at] google.com
 
  
g

GG

     swirl









   
 

 
      






+ Career interest:
    Developing large-scale information retrieval systems that harness Web data such as
    news, blogs, images, videos, social networks and stock quotes.  I am interested in shepherding ideas from
    concepts to the products that people need and enjoy.

                                     

+ Current affiliation (2004 onwards)
     tech lead @ google, in the area of machine learning with applications to 
    ad-targeting
and image/video analysis.  See publications and news section for details. 


    [updates, 11/15/2009]
    We just launched Google Image Swirl, a large-scale content-based Web image organization engine.

+ Past endeavours / other
    
- part owner of matsukei venture (Tokyo)
      
-- business and technical due diligence on start-ups related to online advertising and business analytics.
    - volunteer @ united nations

    - ph.d. candidate @  georgia tech (machine learning, statistical models and computer vision)
   
- grew up in China, Japan and U.S. and speak three languages.


+ Selected Publications, News and Recognition: (Publication lists)
   - Image and video search

        -  A Google Prototype for a Precision Image Search

           New York Times, 2008-04-28

      - VisualRank, applying PageRank to Google image search

          IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) , 2008-09


     - Technology assistance to National Center for Missing and Exploited Children.

          Reuters, BBC, 2008-05


        -
A farewell to Keywords

          Scientific American, 2006-05


 - Social network analysis for [ads-targeting | video recommendation]
        - Swarming the Shelves -- how shops can exploit people's herd mentality to increase sales.

          The Economist, 2006-11-11, p.90 
          Technology developed for dentsu, the most preeminent advertising firm in Japan.


        - Video Suggestion and Discovery for YouTube.

           International World Wide Web Conference (WWW), 2008-04


-  Machine Learning
     
- Boosted Bayesian Network Classifiers

          Machine Learning Journal, 2008-10


        -
Boosted Augmented Naive Bayes

          International Conference on Machine Learning (ICML), 2005 (Best Student Paper)