1 Visual Analytics 1 CS 7450 - Information Visualization November 18, 2013 John Stasko Topic Notes Agenda • Today  Overview of what the term means and how it relates to information visualization  Some example VA research projects • Next time  Specific example, Jigsaw, helping investigative analysis  Related systems Fall 2013 2 CS 74502 3 3 Acknowledgment Slides looking like this provided courtesy of Jim Thomas CS 7450 Visual Analytics • A new term for something that is familiar to all of us • Informal description:  Using visual representations to help make decisions  Sounds like infovis, no? Fall 2013 4 CS 74503 Before There was VA • Growing concern from some that infovis was straying from practical, real world analysis problems • Infovis typically not applied to massive data sets • Infovis “competes” with other computational approaches to data analysis  Statistics, data mining, machine learning Fall 2013 CS 7450 5 Important Paper • Shneiderman suggests combining computational analysis approaches such as data mining with infovis – Discovery tools  Too often viewed as competitors in past  Instead, can complement each other • Each has something valuable to contribute Fall 2013 CS 7450 6 Shneiderman Information Visualization ‘024 Fall 2013 CS 7450 7 Alternatives • Issues influencing the design of discovery tools:  Statistical Algorithms vs. Visual data presentation  Hypothesis testing vs. exploratory data analysis • Pro’s and Con’s? Differing Views • Hypothesis testing  Advocates: By stating hypotheses up front, limit variables and sharpens thinking, more precise measurement  Critics: Too far from reality, initial hypotheses bias toward finding evidence to support it • Exploratory Data Analysis  Advocates: Find the interesting things this way, we now have computational capabilities to do them  Skeptics: Not generalizable, everything is a special case, detecting statistical replationships does not infer cause and effect Fall 2013 CS 7450 85 Fall 2013 CS 7450 9 Recommendations • Integrate data mining and information visualization • Allow users to specify what they are seeking • Recognize that users are situated in a social context • Respect human responsibility Further Questions • Are information visualizations helping with exploratory analysis enough? • Are they attempting to accomplish the right goals? Fall 2013 CS 7450 106 Fall 2013 CS 7450 11 Another Important Paper • Information visualization systems inadequately supported decision making:  Limited Affordances  Predetermined Representations  Decline of Determinism in Decision-Making • “Representational primacy” versus “Analytic primacy”  Telling the truth about your data versus providing analytically useful visualizations Amar & Stasko InfoVis ‘04 Best Paper TVCG ‘05 Task Level • Don’t just help “low-level” tasks  Find, filter, correlate, etc. • Facilitate analytical thinking  Complex decision-making, especially under uncertainty  Learning a domain  Identifying the nature of trends  Predicting the future Fall 2013 CS 7450 127 Fall 2013 CS 7450 13 Analytic Gaps • Analytic gaps – “obstacles faced by visualizations in facilitating higher-level analytic tasks, such as decision making and learning.”  Worldview Gap  Rationale Gap Fall 2013 CS 7450 14 Knowledge Precepts • For narrowing these gaps  Worldview-Based Precepts (“Did we show the right thing to the user?”) Determine Domain Parameters Expose Multivariate Explanation Facilitate Hypothesis Testing  Rationale-Based Precepts (“Will the user believe what they see?”) Expose Uncertainty Concretize Relationships Expose Cause and Effect8 Fall 2013 CS 7450 15 Application of Precepts Fall 2013 CS 7450 16 Application of Precepts9 More Motivation • Increasing occurrences of situations and areas with large data needing better analysis  DNA, microarrays  9/11 security  Business intelligence Fall 2013 CS 7450 17 Articulating the Motivation Fall 2013 CS 7450 18 Video http://videotheque.inria.fr/videotheque/doc/63510 History • 2003-04 Jim Thomas of PNNL, together with colleagues, develops notion of visual analytics • Holds workshops at PNNL and at InfoVis ‘04 to help define a research agenda • Agenda is formalized in book Illuminating the Path, shown on next slide Fall 2013 19 CS 7450 # Visual Analytics Definition Visual analytics is the science of analytical reasoning facilitated by interactive visual interfaces. People use visual analytics tools and techniques to  Synthesize information and derive insight from massive, dynamic, ambiguous, and often conflicting data  Detect the expected and discover the unexpected  Provide timely, defensible, and understandable assessments  Communicate assessment effectively for action. “The beginning of knowledge is the discovery of something we do not understand.” ~Frank Herbert (1920 - 1986) Thomas & Cook 2005 CS 745011 Visual Analytics • Not really an “area” per se  More of an “umbrella” notion • Combines multiple areas or disciplines • Ultimately about using data to improve our knowledge and help make decisions Fall 2013 21 CS 7450 Main Components Interactive visualization Computational analysis Analytical reasoning 22 CS 7450 Fall 201312 Alternate Definition • 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, chapter in Information Visualization: Human-Centered Issues and Perspectives, 2008 Fall 2013 23 CS 7450 Synergy • Combine strengths of both human and electronic data processing  Gives a semi-automated analytical process  Use strengths from each From Keim Fall 2013 24 CS 745013 InfoVis Comparison • Clearly much overlap • Perhaps fair to say that infovis hasn’t always focused on analysis tasks so much and that it doesn’t always include advanced data analysis algorithms  Not a criticism, just not focus  InfoVis has a more narrow scope  (Some of us actually do believe that infovis has/should include those topics) Fall 2013 25 CS 7450 Academic Context Visual Analytics ~2005 Information Visualization ~1990 Pure analytical reasoning Computational analysis Artsy casual infovis, etc. 26 CS