1 / 9

The Interplay Between Mathematics/Computation and Analytics

The Interplay Between Mathematics/Computation and Analytics. Haesun Park Division of Computational Science and Engineering Georgia Institute of Technology FODAVA Review Meeting Dec. 3, 2009. FODAVA and Me. PhD 87 Cornell, numerical comp. (numerical linear algebra),

feoras
Download Presentation

The Interplay Between Mathematics/Computation and Analytics

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. The Interplay Between Mathematics/Computation and Analytics Haesun Park Division of Computational Science and Engineering Georgia Institute of Technology FODAVA Review Meeting Dec. 3, 2009

  2. FODAVA and Me • PhD 87 Cornell, numerical comp. (numerical linear algebra), parallel computing, signal processing, .. • Early 00’s, data analysis, text analysis, bioinformatics – dimension reduction (LDA, NMF), classification, clustering, .. • 2003-2005 NSF Program Director CCF • TF/Numeric, Symbolic, Geometric Computing • Graphics and Vis until Larry Rosenblum • MSPA/MCS (Visual Analytics from Larry R.) • 2005, Georgia Tech, Division of Computational Science and Eng. • 2007, Apr., DIMACS Workshop: Recent Advances in Math. and Information Sciences for Analysis and Understanding of Massive and Diverse Sources of Data , Wen Masters of ONR, and Fred Roberts of DIMACS, • 2007 Fall, FODAVA, on campus team already formed based on CSE Seminar series (J. Stasko, …)

  3. Visual Analytics I see, therefore, I analyze better. Data Representation and Transformation • Visual Analytics is the Science of Analytical Reasoning facilitated by Interactive Visual Interfaces (Thomas and Cook) • Visual Analytics combines automated analysis techniques with interactive visualizations for an effective understanding, reasoning and decision making on the basis of very large and complex data sets (Keim et al.) Analytical Reasoning “Solving a problem simply means representing it so that the solution is obvious.” Herbert Simon, 96 Visual Representation and Interaction Production Presentation and Dissemination

  4. Visual Analytics is Truly Interdisciplinary • Community very broad • Vis, HCI, Database, Cognitive Science, • FODAVA : math, statistics, computational science, data analysis, … • Challenges • Communications • Different communities are used to different problem settings • Opportunities • Maximally utilizing what human and computer can offer • Visual Representation and Interaction • Writable vis (in contrast to readable vis) makes vis useful (P. Hanrahan) • is the key facilitator between human and data

  5. Modules in Data and Visual Analytics System • Data Representation & Transformation Tasks • Classification • Clustering • Regression • Dimension reduction • Density estimation • Retrieval of similar items • Automatic summarization • … • Analytical Reasoning Tasks • Identify the individual leaking classified information • Determine if a set of suspicious events are related • Predict the next stock market crash • Identify best medical treatment based on genomic, population data • … Mathematical , Statistical, and Computational Methods Human Knowledge Visual Representation and Interaction

  6. Challenges • Data Representation and Transformation vs. Analytical Reasoning Tasks • Data representation and transformation concerned with answering questions for which solution process is rather well defined • Analytical reasoning concerned with determining what questions to ask (e.g., formulating hypotheses) • Analysts must continually iterate between these tasks • Evaluation ? • Scalability ? • How does the Interplay come together ? • Visualization is the way for the interplay to occur effectively • Better identification of mapping between existing data representation and transformation and the steps of analytical reasoning tasks needed • Expansion and refinement of data analytical tasks needed for extended mapping • Careful design of visual representation and interaction

  7. Interdisciplinary Activities • Close collaboration of FODAVA teams with people in VA from vis and/or analytical reasoning community critical (ex. J. Stasko, NVAC (J. Thomas, S. Bohn), W. Ribarsky, DHS CoE: David Ebert ) • FODAVA Test bed

  8. Where to publish? • IEEE TPAMI - IEEE Trans. on Pattern Analysis and Machine Intelligence • JMLR – Journal of Machine Learning Research • Information Visualization • InfoVis - IEEE Information Visualization Conference • VAST - IEEE Visual Analytics Science and Technology • NIPS - Neural Information Processing Systems • ICML - International Conference on Machine Learning • SIGKDD - ACM Special Interest Group on Knowledge Disc. and Data Mining • SDM - SIAM Conference on Data Mining • ICDM - IEEE International Conference on Data Mining • CVPR - Conference on Computer Vision and Pattern Recognition • WWW - International World Wide Web Conference • WSDM - International Conference on Web Search and Data Mining • AISTATS - International Conference on Artificial Intelligence and Statistics • CompStat - International Conference on Computational Statistics • JSM - (American Statistical Association's) Joint Statistical Meetings

More Related