Research Visual Analytics Visualization Research Work Silvia Miksch Silvia Miksch: Short History Linz
Backbone of Research Project Ecosystem www.cvast.tuwien.ac.at Content Part 1: Visual Computing Doctoral College Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding EXPAND Visual Computing Conclusion Content Part 1: Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding Visualization Success Story [Tufte, 1997] adapted from [Hearst, 2004] Mystery: What is causing a cholera epidemic in London in 1854? Conclusion
London 1854 9
+ * + * * to do 1.... 2.... 3.... to do 1.... 2.... 3.... to do 1.... 2.... 3.... + * + * * to do 1.... 2.... 3.... to do 1.... 2.... 3.... to do 1.... 2.... 3.... + * + * * to do 1.... 2.... 3.... to do 1.... 2.... 3.... to do 1.... 2.... 3.... Which Information to Tackle...... Change over Time Data Patient Data * * * Guidelines + + Patient Data Patient Data Patient Data Guidelines Guidelines Guidelines Users Tasks to do 1.... to do 2. 1.... to do 3.... 2. 1.... 3.... 2.... 3.... time Data & Information Big Data time-oriented, multivariate, irregular sampled, having different temporal granularities, qualitative, quantitative, etc. structured and unstructured enriched by meta data Motivation: Main Problems Data Unmanageable Information Overload variables On the one hand, a huge amount of highly structured data and information is available in working situations and the daily life,... On the other hand, different kinds of data and information analysis methods were developed to gain more insights (information and knowledge gains). Missing Integration of Various (Heterogeneous) Information Sources Various Interdisciplinary Methods time Missing Involvement of Users and their Tasks
Visualization for Problem Solving Analytical Methods 1 599 693 2 525 693 3 541 662 4 542 611 5 527 579 6 505 529 7 469 553 8 409 558 9 321 531 10 318 606 11 321 693 12 243 693 13 250 660 14 253 579 15 246 527 16 230 489 17 196 510 18 192 497 19 134 508 20 94 493 21 25 423 22 87 467 23 128 482 24 183 473 25 163 568 26 541 558 27 542 531 28 527 606 29 505 693 30 469 693 31 409 660 32 321 579 33 318 527 34 321 489 35 243 510 36 250 497 37 253 508 38 246 493 39 230 423 40 196 467 41 192 482 42 134 473 43 94 568 44 541 558 45 542 531 46 527 606 47 505 693 48 469 693 49 409 660 50 321 579 51 318 527 52 321 489 53 243 510 54 250 497 55 253 508 56 246 493 57 230 423 58 196 467 59 192 482 60 134 473 61 94 568 62 541 558 63 542 531 64 527 606 65 505 693 66 469 693 67 409 660 68 321 579 69 318 527 70 321 489 71 243 510 72 250 497 73 253 508 74 246 493 75 230 423 76 196 467 77 192 482 78 134 473 79 94 568 80 318 558 81 321 531 82 243 606 83 250 693 84 253 693 85 246 660 86 230 579 87 196 527 88 318 489 89 321 510 90 243 497 91 250 508 92 253 493 93 246 423 94 230 467 95 196 482 96 318 473 97 321 568 98 243 482 99 250 473 100 253 568 101 246 606 102 230 693 103 196 693 104 230 660 105 196 579 106 318 527 107 321 489 108 243 606 109 250 693 110 253 693 111 246 660 112 230 579 113 230 527 114 196 489 115 318 510 116 321 497 117 243 508 118 250 493 119 253 423 120 246 467 Screen Resolution: 1024 * 768 = 786.432 Yearly Measurements of Water Level in Low.Austria: 1 5.256.000 Number of Cellular Phones in Austria (2005): 2 8.160.000 Transmitted Emails Every Hours (World-Wide): 3 35.388.000 Whole Data often not Presentable 1. Applying Analytical Methods (Data Reduction) 2. Visualization of Most Important Data and Information Analytical Methods Statistics, Machine Learning & Data Mining 1... Amt der NÖ Landesregierung, Abt. WA5 - Hydrologie, http://www.noel.gv.at/service/wa/wa5/htm/wnd.htm 2... CIA Factbook, https://www.cia.gov/cia/publications/factbook/ 3... How Much Information?, UC Berkeley, http://www2.sims.berkeley.edu/research/projects/how-much-info-2003/ today: peta (10 15 ) tomorrow: exa (10 18 )& zeta (10 21 ) Interactions Visual Analytics Past Today Only passive Observations Representation not Changeable one fits all Active Examination with Visualizations Dynamically Adaptable and Modifiable Different Users, Tasks, and Aims James Thomas & Kristin A. Cook NVAC (National Visualization and Analytics Center), Seattle, USA Visual Analytics is the science of analytical reasoning facilitated by interactive visual interfaces [Thomas & Cook 2005] [Keim, et al. 2010]
Visual Analytics Process [Keim, et al., 2008] Knowledge Generation Model for VA [Sacha, et al., 2014] Knowledge Generation Model for VA [Sacha, et al., 2014] Time has a Complex Structure
Visual Analytics of Time-Oriented Data User-Centered Design data characterizing time & time-oriented data visualizing time-oriented data interacting with time analyzing time-oriented data goals/tasks Interactive Visual Analytics Methods appropriateness user/audience Content Part 1: Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding Conclusion
CareCruiser Project Interactive Exploration of Effects of Therapeutic Actions on a Patient s Condition Wolfgang Aigner, Theresia Gschwandtner, Katharina Kaiser, Silvia Miksch, Andreas Seyfang [Gschwandtner, et al., 2010-2011] CareCruiser Users Medical Experts & Physicians Data Patient Data and Treatment Plans Data: multivariate, abstract Time: linear, instant Task Exploring the Effects of Clinical Actions on a Patient s Condition [Gschwandtner, et al., 2010, 2011] Communicating Different Aspects with Multiple Views Communicating Different Aspects with Multiple Views
Communicating Different Aspects with Multiple Views Communicating Different Aspects with Multiple Views Visualizing Temporal Aspects Visualizing temporal aspects Treatment Plans and Patient Data Patient Parameter Chart Treatment Plans and Patient Data Patient Parameter Chart: Possible Value Range maximum value mimimum value
Visualizing temporal aspects Aligning Treatment Plans or Clinical Actions Treatment Plans and Patient Data Patient Parameter Chart: Intended Value Range intended value range 38 Interacting with Time CareCruiser Video [Gschwandtner, et al. : CareCruiser]
Interacting with Time [Gschwandtner, et al. : CareCruiser] Lessons Learned Visual Analytics: patient data in combination with applied treatment plans and clinical actions Step-wise interactive exploration of effects New insights Generation of hypotheses Evaluation (different points of view): Collaboration with medical expert guided design Heuristic Usability evaluation technical point of view Case study domain point of view [Lammarsch, et al., 2011, 2013], [Bögl, et al., 2011, 2013] Modeling Hypotheses with Visual Analytics Methods to Analyze the Past and Forecast the Future Wolfgang Aigner, Markus Bögl, Tim Lammarsch, Silvia Miksch, Alexander Rind Peter Filzmoser TiMoVA Visual Analytics for Model Selection in Time Series Analysis Users Experts in time series analysis Data Data: univariate, abstract Time: instant Tasks Time series transformation and model selection
Example: Statistical SW Tool Gretl Preview-Video [Bögl, et al., 2011, 2013] Usage Scenario Model Selection TiMoVA VA Prototype Time Series Line Plot [Bögl, et al., 2011, 2013] Definition of [Bögl, et al., 2011, 2013] Autocorrelation Function (ACF) Residual Analysis [Shumway and Stoffer, 2011]
TiMoVA VA Prototype [Bögl, et al., 2011, 2013] Usage Scenario Model Selection Time Series Line Plot [Bögl, et al., 2011, 2013] Autocorrelation Function (ACF) Residual Analysis Lessons Learned Visual Analytics of Time Oriented Data Evaluation Applied usage scenarios based on the requirements Formative evaluation during design and implementation phase Demonstration session internally and with two external experts Supports and guides domain experts by Model order selection inside the relevant plot Immediate visual feedback of the model residuals Visualization of the model transitions Short visual feedback cycles
Visual Analytics of Time-Oriented Data Information Discovery Process Usability & Utility Insights Studies Knowledge Respositories Infrastructure for Reusable Components TimeBench: A Data Model and Software Library Infrastructure to Facilitate Evaluation EvalBench: A Software Library for Visualization Evaluation [Rind, et al., 2013] Multiple Granularities Different Time Primitives Temporal Indeterminacy Science of Interaction Role and Value of Interactivity Supported by the Austrian BMWFJ via CVAST, a Laura Bassi Centre of Excellence (#822746), and by the Austrian Science Fund (FWF) via HypoVis (#P22883) [Aigner, et al., 2013] Reduces implementation effort for evaluation features Consistent and reproducible execution of study protocols Integrates well with existing visualization prototypes Free and open source software (@GitHub) How to use? [Aigner, et al., 2013] Supports: Controlled Experiments Interaction Logging Laboratory Questionnaires Heuristic Evaluations Insight Diaries
How to use? [Aigner, et al., 2013] How to use? [Aigner, et al., 2013] Define task lists for sessions Define task lists for sessions Implement EvaluationDelegate interface How to use? [Aigner, et al., 2013] Content Part 1: Define task lists for sessions Implement EvaluationDelegate interface Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding Conclusion
Challenges (Time-Oriented) Data Scale and Complexity Heterogeneous Data (Meta Data, Semantics, Multiple Sources, etc.) Data Quality & Uncertainty Data Provenance Application Challenges in Visual Analytics User Meeting Users Needs High Degree of Interactivity (Temporal Dimensions) Evaluation (Qualitative & Quantitative) Challenges (Time-Oriented) Data Scale and Complexity Heterogeneous Data (Meta Data, Semantics, Multiple Sources, etc.) Data Quality & Uncertainty Data Provenance Application Challenges in Visual Analytics User Meeting Users Needs High Degree of Interactivity (Temporal Dimensions) Evaluation (Qualitative & Quantitative) Design Technology Design Technology Guidance on how to Design and Develop Visual Analytics Systems Provide Reusable Infrastructure Guidance on how to Design and Develop Visual Analytics Systems Provide Reusable Infrastructure [VisMaster Challenges 2010] [VisMaster Challenges 2010] Challenges: Visual Analytics Process [Keim, et al., 2008] Conclusion Visual Analytics Detect the Expected and Discover the Unexpected data goals/tasks Interactive Visual Analytics Methods appropriateness user/audience
Thanks to Alan Albert Alessio Alexander Alexander Alime Amin Andreas Andreas Annette Arghad Barbara Barbara Ben Bilal Brain Burcu Carlo Catherine Christian Christian Christian Claudio Daniel David Dorna Edeltraud Eduard Elisabeth Elpida Elske Eva Fabian Felix Florian Florian Frank Franz Gennady Georg Georg Gerhard Gerhilde Guiseppe Hanna Heidrun Helga Helwig Ingrid Jarke Jim Jimmy Johannes Jörn Jürgen Kai Karl Katharina Klaus Krist Luca Lukas Manfred Mar Margit Maria Markus Markus Martin Martin Matt Michael Michael Michael Mikko Monika Monika Mor Nada Natalie Nikolaus Otto Panagiotis Paolo Paolo Patrick Peter Peter Peter Rene Rita Robert Robert Robert Roberto Ruth Sabine Salvo Samson Silvana Simone Sophie Stefan Stefan Stephan Susanne Sylvia Taowei David Theresia Thomas Tim Tom Werner Wolfgang Yuval... and many students and co-workers Wolfgang Aigner Silvia Miksch Heidrun Schumann Christian Tominski Visualization of Time-Oriented Data with a foreword by Ben Shneiderman Springer 1st Edition., 2011, XVIII, 286 p. 221 illus., 198 in color. Hardcover, ISBN 978-0-85729-078-6 Due: June 2011 Table of Contents Introduction Historical Background Time & Time-Oriented Data Visualization Aspects Interaction Support Analytical Support Survey of Visualization Techniques Conclusion www.timeviz.net www.timeviz.net Horizon Graph [Reijner, 2005]
Content Part 1: Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: Part 2 HOW DOES RESEARCH WORK How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding Conclusion [Hevner et al. 2004] WARUM habe ich WAS gemacht, WIE habe ich es gemacht und mit WELCHEM ERGEBNIS. 7
[Hevner et al. 2004] [Hevner et al. 2004] Research & Teaching Bachelor Course Publications Conferences Journals Master Courses Supervision of Bachelor Students Supervision of Master Students Contributions in &Books Supervision of PhD Students
>> Title, Affiliation << [Simon L. Peyton Jones, 2004 presentation] >> References << Peer Reviewing Process Roles Different in each community Different... PC IPC Different for Journals Conferenes Timely and Complex Process Steps... Example EuroVis 2012 & VisWeek 2012 7 20.4.2 188.917 VU Informationsdesign und Visualisierung 8
Primary and Secondary Reviewers 8 8 Review Forms Review Forms 8 8
Review Forms Review Forms 8 8 Review Forms Ethics Guidelines Be Timely Protect Ideas Avoid Conflict of Interest Be Specific Be Helpful Be Tactful (In Summary) http://www.uib.no/eurovis2011/reviewing_guidelines.php http://vgtc.org/wpmu/techcom/conferences/ethics-guidelines/#in%20summary 8 8
Funding Backbone of Research Project Ecosystem www.cvast.tuwien.ac.at Research needs funding...... Basic Research Doctoral College... Applied/Coooerative Reserch Visual Computing EXPAND Visual Computing Thanks to (Intern)ational Collaborations Alan Albert Alessio Alexander Alexander Alime Amin Andreas Andreas Annette Arghad Barbara Barbara Ben Bilal Brain Burcu Carlo Catherine Christian Christian Christian Claudio Daniel David Dorna Edeltraud Eduard Elisabeth Elpida Elske Eva Fabian Felix Florian Florian Frank Franz Gennady Georg Georg Gerhard Gerhilde Guiseppe Hanna Heidrun Helga Helwig Ingrid Jarke Jim Jimmy Johannes Jörn Jürgen Kai Karl Katharina Klaus Krist Luca Lukas Manfred Mar Margit Maria Markus Markus Martin Martin Matt Michael Michael Michael Mikko Monika Monika Mor Nada Natalie Nikolaus Otto Panagiotis Paolo Paolo Patrick Peter Peter Peter Rene Rita Robert Robert Robert Roberto Ruth Sabine Salvo Samson Silvana Simone Sophie Stefan Stefan Stephan Susanne Sylvia Taowei David Theresia Thomas Tim Tom Werner Wolfgang Yuval... and many students and co-workers Content Part 1: Motivation & Contextualization Visual Analytics of Time-Oriented Data Challenges & Opportunities Part 2: How Does Research Work Research &Teaching Scientific Writing Scientific Reviewing Funding Conclusion
TIPS: How to Do Research http://www.ifs.tuwien.ac.at/~silvia/research-tips Part 3 TIPS & TRICKS 9 TIPS: How to Do Research TIPS: How to Do Research
TIPS: How to Do Research TIPS: How to Do Research siehe TIPS: How to Do Research
My Favorite Book for years... Lyn Dupré: BUGS in Writing: a Guide to Debugging Your Prose, Addison-Wesley, 1998 1... and William Strunk Jr., E. B. White The Elements of Style Longman, New York; 4th edition 1999 WARUM habe ich WAS gemacht, WIE habe ich es gemacht und mit WELCHEM ERGEBNIS. 1 1