2008
-
Stephen Kobourov
joins as Principal Research Staff Member
Stephen's research interests include information visualization, graph
drawing and geometric algorithms, emphasizing problems relating to graph
visualization. He received a BS in Computer Science and Mathematics
from Dartmouth College in 1995 and PhD in Computer Science from Johns
Hopkins University in 2000. He worked at the University of Arizona where
he received a NSF Career grant and was tenured in 2006. As a Fulbright
Scholar he spent a sabbatical year at the University of Botswana. He
is an editor of the Journal of Graph Algorithms and Applications and
has served on program committees for SODA, ESA, GD and SoftVis, and as
program committee chair for the 10th Symposium on Graph Drawing.
-
Summer 2008 visitors:
- A paper on a multidimensional long term use study conducted with
LiveRAC,
co-authored with
Peter McLachlan and
Tamara Munzner from
the University of British Columbia, with Stephen North
and Eleftherios Koutsofios, was presented at
ACM CHI 2008
in Florence, Italy, in April. This was based on a
field study of a tool for scalable/stretchable views of
large sets of time series, conducted with AT&T life
cycle engineers in the Internet Hosting business.
|
The National Academy of Television Arts and Sciences awarded
AT&T an Emmy for the
invention of coax cable. Our boss,
DGB,
accepted the award on Jan 7. The transistor could have a shot at an award, too,
but the Academy is waiting to see some impact outside the lab. If you visit us,
we'll take your photo with the Emmy.
|
- Yifan Hu is working on fast large graph layout and
has created a
gallery
worth checking out. Samples below. Watch this space.
2007
-
Congratulations to Yehuda Koren, who led
Team BellKor
in winning the
Netflix
Competition Progress Prize. The other team members
were Bob Bell and Chris Volinsky. In addition to the above,
Netflix also sent us our very own
star
shown to the right. We have it in the lab. It weighs 200 lbs.
An article on the statistical and computational methods they developed was published in the December 2007 issue of Statistical Computing and Graphics newsletter.
Yehuda gave two presentations at KDD 2007:
Modeling Relationships at Multiple Scales to Improve Accuracy of Large
Recommender Systems with Chris Volinsky and Robert Bell,
and Improved Neighborhood-based Collaborative Filtering
with Robert Bell (in the KDD-Cup workshop). He co-authored
a third paper, Fast Direction-Aware Proximity for Graph Mining,
with Hang-Hang Tong and Christos Faloutsos from CMU. This was
an extension of last year's work on Proximity Subgraphs.
-
The 2.18 release of Graphviz
is out. The Windows version now supports most of the recent improvements
like Cairo rendering, the gvpr stream processor and custom shapes,
and it's faster, too.
-
We set up a new minilab to study the integration of 3d computer
graphics and video, especially in realtime. It has a custom
frame for mounting HDTV cameras that connect over gigabit LAN
to a dedicated cluster with ATI graphics cards and adequate
local and centralized storage. Some projects include multipoint
video capture, novel view synthesis, gaze-corrected video conferencing,
camera motion estimation and surface geometry estimation.
The minilab also benefits from the acquisition of a second
graphics cluster to drive a 32-LCD powerwall under development,
as part of a broader 768-core multicluster project in AT&T Labs.
-
Information Visualization Research and AT&T Labs researcher
Bill Cheswick
collaborated with
Lumeta Corp. and
Matt Dresdner at Mercury Seattle to make
artwork for an Internet map poster. Click on the image at
the right to download the poster (warning: 3.2 meg PDF file).
|
|
-
What's the connection between Doris Day and Meryl Streep?
We developed new algorithms that quantify ``proximity'' or closeness
in social networks and other quasi-random graphs, and presents
a network diagram that shows the result.
Click
here to try your own queries in the
IMDB movie database or the
DBLP computer science bilbiography.
|
|
-
We are very happy to announce that
Yifan Hu
joined our group in September.
Yifan is an expert in network analysis, optimization, and large-scale
network visualization. He recently worked at Wolfram Research and
in high performance computing at the Daresbury Laboratory in Cheshire, UK.
|
