http://sna-hackathon.tumblr.com/
>>> Social Network Big Data Hackathon <<<
at the INSNA Sunbelt Conference
March 13-14, 2012
Crowne Plaza Hotel, Redondo Beach, CA
Manual collection of social network data is so XXth century! New
sources of social network data such as Twitter, Facebook,
OpenCongress, CrunchBase, and many others, present data at the scale
up to 100s of millions of nodes and billions of edges, with millions
of nodes and edges added every hour.
In this environment, traditional SNA algorithms and metrics falter due
to their high computational complexity and difficulty of
parallelization. Big Data approaches such as Map/Reduce, Hive, Pig and
Storm, distributed data storage solutions of Hadoop, Redis and MongoDB
hold a promise that SNA algorithms can be made to work over web-scale
data — but much work is required to make this happen.
We are soliciting papers, demos and tutorials on:
* Software — open-source or academic-license tools
* Big Social Data problems, solutions and war stories
* Data warehousing, data manipulation, data wrangling and wrestling
* Streaming graph analysis, real-time data analysis, dynamic network analysis
* Real-world applications
* … any topic relevant to analysis of big, real-time, social network
and social stream data
Papers from the industry and academia are welcome.
We extend a special welcome to companies fueling this movement — if
you work at Twitter, Facebook, Bit.ly, Instagr.am, Tumblr, etc.,
please consider coming to present, teach and learn! Please bring your
data and your open-source tools to share!
Part 1 of the workshop shall be devoted to practical aspects of
dealing with big data — perhaps more of a tutorial to get everyone up
to speed. We’ll select practical papers for this section.
Part 2 will be in a Hackathon format — we will start by setting up a
distributed, real-time cloud computing “playground”, turn on a fast
and high-volume data stream, and, at the end of the workshop, leave
with meaningful adaptations of SNA metrics and algorithms to
real-time, high-volume data. We will encourage formation of impromptu
teams and collaborations, especially joint work involving academics
and industry.
Part 3 — The Red-Eye session — will be devoted to presentations of the
hackathon participants and academic and industry “war stories”.
Hackathon attendees are expected to be proficient programmers in
languages of their choice, and bring high-performance laptops to the
workshop. They’ll need all the computing horsepower they can get.
We’ll also provide generic Amazon EC2 instances for cloud computing
Submission guidelines:
* Please submit abstracts (300-500 words) to [log in to unmask]
* For software demos, please include a URL to your site or
documentation repository or GitHub (Google Code, etc) open-source
repository.
Deadline: January 15, 2012
|