SAMSI-FODAVA Workshop on Interactive Visualization and Analysis of Massive Data – December 10-12, 2012


With the advance in technology, enormous amounts of data are generated on a daily basis virtually in every area including bioinformatics, astrophysics, chemometrics, social network analysis, web mining, text mining, financial analysis, and security. We are faced with significant analytical challenges due to many special characteristics of these data sets, which are of large volume, often unstructured, high dimensional, noisy, incomplete, time-varying, spatial, and originate from different sources. These challenges can be turned into new opportunities and discoveries when the massive data can be transformed into useful knowledge. Recent developments in data and visual analytics show that incorporating interactive capability through visual interfaces with automated data analysis methods can substantially increase our ability to understand the data and find more meaningful solutions.

The primary goal of the workshop was to bring together researchers in Mathematics, Statistics, Computational Science and Engineering, Computer Science, and Visualization to work on massive scale data and visual analytics. Issues that were investigated include the mathematical, statistical, and algorithmic issues in efficient representation and transformation of data, scalable and dynamic algorithms for real time interaction, visual representation in limited screen space, performing evaluations, and applications.

Schedule and Supporting Media

Monday, December 10
at SAMSI, Room 150

Time Description Speaker Slides Videos
9:00-9:30 Registration and Continental Breakfast
9:30-9:40 Welcome Remarks
9:40-10:10 The Role of Visualization and Analytics in Solving Problems Based on Massive Data Daniel Keim, University of Konstanz
10:10-10:40 Robust Subspace Modeling Gilad Lerman, University of Minnesota  
10:40-11:10 Break
11:10-11:40 On the Complexity of Statistical Algorithms Santosh Vempala, Georgia Tech
11:40-1:40 Lunch
1:40-2:10 Interactive Graphics for Data Exploration Heike Hofmann, Iowa State University
2:10-2:40 The Role of Perception in Visualization and Visual Analytics Christopher Healey, North Carolina State University
2:40-3:10 Break
3:10-3:40 Visual Analytics for Evidence-Based Medicine David Gotz, IBM
3:40-4:10 Student Poster Fast Forward
4:10-5:00 Break
5:00-7:00 Poster Session and Reception

Tuesday, December 11
at SAMSI, Room 150

Time Description Speaker Slides Videos
9:00-9:30 Registration and Continental Breakfast
9:30-10:00 Large-Scale Visual Data Analysis Chris Johnson, University of Utah
10:00-10:30 Computational Signal Processing in Smart Patient monitoring: Algorithms, Applications and Future Challenges Sabine Van Huffel, K.U. Leuven, ESAT/SISTA
10:30-11:00 Break
11:00-11:30 Discovery of Mechanisms and Prognosis of Cancers from Matrix and Tensor Modeling of Large-Scale Molecular Biological Data Orly Alter, University of Utah  
11:30-Noon The Combinatorial Laplacian and Dimension Reduction Sayan Mukherjee, Duke University
Noon-2:00 Lunch
2:00-2:30 TB-Vis: Visualizing TB Patient-Pathogen Relationships Kristen Bennett, RPI
2:30-3:00 VisIRR: Visual Information Retrieval and Recommendation System for Document Discovery Alexander Gray, Georgia Tech
3:00-3:30 Break
3:00-4:30 Panel Chair:
Jimeng Sun, IBM
Polo Chau, Georgia Tech;
Daniel Keim, University of Konstanz;
Larry Rosenbaum, National Science Foundation;
Leland Wilkinson, SYSTAT/Northwestern University
6:00 Dinner on your own in small groups at Southpoint mall
Radisson hotel van service will take us to Southpoint mall

Wednesday, December 12
at SAMSI, Room 150

Time Description Speaker Slides Videos
9:00-9:30 Registration and Continental Breakfast
9:30-10:00 New Approaches for Nonlinear Dimensionality Reduction Fei Sha, University of Southern California
10:00-10:30 New Approaches to Storytelling from Massive Textual Datasets Naren Ramakrishnan, Virginia Tech
10:30-11:00 Break
11:00-11:30 Scalable Bayesian Learning for Matrix and Tensors Alan Qi, Purdue University
11:30-Noon BigData: Probabilistic Methods for Efficient Search and Statistical Learning in Extremely High-Dimensional Data Ping Li, Cornell University
Noon-12:10 Concluding Remarks
12:10-1:30 Lunch