[san-tech][03111] Re:"Government Developing Data Mining Tools To Fight Terrorism", 2011/05/12

Date: Tue, 24 May 2011 20:21:52 +0900
--------------------------------------------------
2011/06/11
[san-tech][03129] Re:"Government Developing Data Mining Tools To Fight Terrorism", 2011/05/12
--------------------------------------------------
[san-tech][03087] "Government Developing Data Mining Tools To Fight Terrorism", 2011/05/12

Cray XMTが導入されてる DOE PNNL (Pacific Northwest National Laboratory)

Center for Adaptive Supercomputing Software-Multithreaded Architectures
(CASS-MT)
  http://cass-mt.pnnl.gov/

と共同研究をしている、Georgia Institute of Technologyの David A. Bader
教授の国際会議での発表です：



"Massive Social Network Analysis: Mining Twitter for Social Good"
 David Ediger, Karl Jiang, Jason Riedy, David A. Bader, Courtney Corley,
 Rob Farber and William N. Reynolds
 The 39th International Conference on Parallel Processing (ICPP 2010),
 San Diego, CA, September 13-16, 2010.
  http://www.cc.gatech.edu/~bader/papers/MassiveTwitter.html
  http://www.cc.gatech.edu/~bader/papers/MassiveTwitter-ICPP2010.pdf

  "Social networks produce an enormous quantity of data. Facebook
   consists of over 400 million active users sharing over 5 billion
   pieces of information each month. Analyzing this vast quantity of
   unstructured data presents challenges for software and hardware.
   We present GraphCT, a Graph Characterization Toolkit for massive
   graphs representing social network data. ..... "We use GraphCT to
   analyze public data from Twitter, a microblogging network. Twitter's
   message connections appear primarily tree-structured as a news
   dissemination system." ....

GraphCT: a Graph Characterization Toolkit for the Cray XMT
  http://www.cc.gatech.edu/~bader/code.html

  "GraphCT is a parallel toolkit for analyzing massive graphs on the
   massively multithreaded Cray XMT. Designed from the ground up to
   take advantage of the large shared memory and fine-grained
   synchronization of the Cray XMT, GraphCT provides an optimized
   library of functions including clustering coefficients, connected
   components, betweenness centrality, graph diameter, and distribution
   statistics."

David A. Bader, Professor
 College of Computing
 Georgia Tech
  http://www.cc.gatech.edu/~bader/