김현지

Contact Information   

    
    Email     :   hjkim@dblab.postech.ac.kr

    Address  :   Dept. of Creative IT Convergence Engineering, POSTECH, Pohang, South Korea

  Education


    Ph.D.           2016.2    –                                             POSTECH, South Korea   

                                                                                      Dept. of Creative IT Engineering

     B.S.             2012.2    –   2016.2                                  POSTECH, South Korea

                                                                                      Dept. of Creative IT Engineering                                                                                                                                                                     

                         2015.4  –   2015.9                               Technische Universität Berlin, Germany                                                                                                           

                                                                                      (exchange student)

  Industry Experience

                                                                                               

     Intern          2015.12    ~   2016.03                                Internship, Winter, 2015-2016             - Efficient graph partitioning on a distributed graph engine

                                                                                      Oracle Labs, Belmont, CA, USA            - Performance analysis and tuning of graph analytics and

                                                                                                            graph queries in large-scale graphs on a distributed graph engine

  Publications

- Kim, H., Lee, J., Bhowmick, S., Han, W., Lee, J., Ko, S., and Jarrah, M., "DualSim: Parallel Subgraph Enumeration in a Massive Graph on a Single Machine," In Proc. 42nd Int'l conf. on Management of Data, ACM SIGMOD, San Francisco, USA, June 2016 (to appear).


  Selected Projects

-      Development of disk-based gradient descendant method (course project)

  In this project, I have developed a disk-based, gradient descendant algorithm on top of TurboGraph, which can be used for disk-based recommendation. During this project, I have analyzed important parts of source code of TurboGraph.


  Areas of Interest

      
      Graph Databases, Graph Data Analysis, Big Data Processing

  Other Links