IEEE VIS 2014 http://ieeevis.org ************************************************************************** VIS'14 Capstone Speaker: Barbara Tversky (Stanford/Columbia) ************************************************************************** Workshop on Visualization for Predictive Analytics: Call For Participation Co-Located with IEEE VIS'14 | Paris 9-14 Nov, 2014 http://predictive-workshop.github.io/ SUBMISSION: Sep. 15, 2014 PARTICIPATION: The workshop is open to all VIS 2014 attendees FOR INQUIRIES: predva.workshop@gmail.com What is the use of visualization in prediction? How do people use visualization in predictive modeling? How can we improve the state of the art in predictive visual analytics? We invite the submissions of *short technical or position papers* (between 1-4 pages) to explore the use of visualization in prediction and showcase existing research. TECHNICAL PAPERS - May include, but are not limited to the following: Novel Visual Analytics or Visualization Techniques * How to assist predictive development, evaluation and communication * Visual analytics techniques for predictive models such as regression and classification * Visual analytics techniques for clustering * Visual analytics techniques for high-dimensional data * Interactive visualization for refining predictive models Applications and Case Studies * Real-world problems and experiences from public sectors and industry * Predictive visual analytics in business, technology, healthcare, finance, telecommunications, etc. * Predictive visual analytics applications in public sectors such as government, development, security, etc. Theory/Methods/Modeling/Evaluation * Real and synthetic data sets and benchmarks * Taxonomies of predictive tasks in visualization * Evaluation and testing in predictive visual analytics POSITION PAPERS - May include, but are not limited to, visionary ideas addressing topics such as: * What's the use of visualization in prediction tasks? * How can visualization help data scientists make sense of predictive models? * How can end users be sure predictive models are made of domain-relevant features? * How can visual analytics help researchers include their domain knowledge into the modeling process? * How do we build benchmark datasets and ground truth to objectively compare different predictive model visualizations? * What are good and suitable processes to ensure usefulness of predictive models? * How can we use visual analytics to “explain” patterns derived from predictive models? * How can we use predictive models as discovery tools? * How can visualization help modelers gain trust into their results? * Is visualization needed at all? * Where is the boundary between predictive, explanatory and exploratory analysis? See http://predictive-workshop.github.io/ for more details. Organizers: Adam Perer, Enrico Bertini, Ross Maciejewski, Jimeng Sun -- Liz -- G. Elisabeta Marai Robotics Institute School of Computer Science Carnegie Mellon University http://visualizlab.org/people/marai