1 / 22

Transfer Viva

Transfer Viva . Empathic Visualisation Algorithm (EVA). Outline. Multi-dimensional Visualisation Overview of EVA Achievements so far What’s happening now Future Work. Data. Large Multiple dimensions (>3) Non-physical nature Hidden information

blade
Download Presentation

Transfer Viva

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. Transfer Viva Empathic Visualisation Algorithm (EVA)

  2. Outline • Multi-dimensional Visualisation • Overview of EVA • Achievements so far • What’s happening now • Future Work

  3. Data • Large • Multiple dimensions (>3) • Non-physical nature • Hidden information • Quantitative (or transformed to quantitative)

  4. Types of Variables • Nominal (equal or not equal) • Ordinal (obeys a < relation) • Quantitative (can do arithmetic)

  5. Taxonomy (1) (a) (b) • Mapping (1) • Arbitrary (a) • Automatic (b) • Visual Structure (2) • Abstract (c) • Naturalistic (d) (2) (c) (d) ?

  6. Advantages Actual Values Quantitative analysis Variables treated uniformly Objective visual structure Disadvantages Hard to get overview User learning time Time to make decisions Not generic Hard to find relationships Complexity increases with dimensionality Arbitrary Mapping - Abstract Visual Structure

  7. Advantages Generic system Actual Values All variables treated equally Less user learning time Disadvantages Time to make decisions hard to get overview hard to find relationships Interactivity Complex processing, and more complicated visual structure Automatic Mapping - Abstract Visual Structure

  8. Advantages Holistic view Simple, easy to use visual structure Less time to make decisions Relationships of the variables Disadvantages Variables not treated uniformly Subjectiveness of visual structure User learning time No actual quantities, values No extreme cases Arbitrary Mapping - Naturalistic Visual Structure

  9. Automatic Mapping - Naturalistic Visual Structure: EVA • interrelationships of data variables, encapsulated in one visual structure • Gross information about the data (overall view) • Simple, easy to understand visual structure • Learning time minimised • Background of users irrelevant • May enable decisions on the fly

  10. ... • Generic system • Complexity of visual structure doesn’t increase with dimensionality

  11. From Data to Naturalistic Visual Structure an Automatic Mapping

  12. Fundamentals of method • n*k data matrix X (row individual, k observations) • Objective: • salient features • overall view • Naturalistic, Automatic • “Value System” • Goal of method: Visual Homomorphism

  13. Assumptions Be p functions over the data - “value system” Be p characteristics of visual structure. Measuring the totality of the visual structure The r features of the visual structure

  14. Assumptions 2 Be feature functions over the data matrix determining the visual structure If and Then we have to choose Such that is minimised

  15. Using Genetic Programming • Minimisation problem can be tackled with a GP • Large collection of random functions - Population • Fitness: The distance measurement

  16. Summary • User identifies the “value system” • System decides on the visual structure and its features • System identifies the characteristics of the visual structure • Fitness function is defined • GP parameters, and run

  17. Achievements so far • Literature survey on Information Visualisation • GP survey • System implemented • Successfully tested using circles • users needed no learning time • users made decisions immediately • users noticed small changes

  18. Conclusions • Can subjects extract information from the visual representation of the data set? Yes • Can this visualisation method act as an aid to the decision making process? Yes

  19. Under process • Experiment using faces • Face: the epitome of a naturalistic visual structure • At final stage

  20. Future Work • For current experiment, test method for: • generalisability • convergence (experimentally) • Final experiment (also using faces) • different application (data set) • divide users to experts/non_experts • test it with different number of characteristics • Measure time to make decisions • Test ‘readability’ of visual structure

  21. Time Plan • Finish the first test by end of January • Finish the main, final experiment by beginning of July • Write up

More Related