ICS 280: Information Visualization 1 ICS 280: Advanced Topics in Information Visualization Professor Alfred Kobsa Overview of Information Visualization

ICS 280: Information Visualization 1

ICS 280: Advanced Topics in Information VisualizationProfessor Alfred Kobsa

Overview of Information Visualization II

April 12, 2001

Víctor González ([email protected])


Definitions (1)

• External CognitionUse of external world to accomplish cognition.Cognitive artifacts e.g. Post-it notes, bookmarks, wedding ring.

• Information designDesign of external representations to amplify cognition.e.g. Maps.

• Data graphicsUse of abstract, nonrepresentational visual representations of data toamplify cognition.



Definitions (2)

• VisualizationUse of computer-based, interactive visual representations of data to amplify cognition

• Scientific visualizationUse of interactive visual representations of scientific data, typically physically based, to amplify cognition.

• Information visualizationUse of interactive visual representation of abstract, nonphysically based data to amplify cognition.e.g. Business information. * large amount of data* interactivity



How Visualization Amplifies Cognition (1)

Classic study of Larkin and Simon (1987)

Solving physics problems using diagrams vs using non-diagrammatic representations. They compared the effort to do search, recognition, and inference.

(1) By grouping together information that is used together large amounts of search were avoided. (2) By using location to group information about a single element, the need to match symbolic labels was avoided, leading to reductions in search and working memory.(3) The visual representation automatically supported a large number of perceptual inferences that were extremely easy for humans. e.g. Recognize geometric elements like alternate interior angles




(1) Increased Resources- Expanded working memory available for solving a problem °- Expanded storage of information - quickly accessible massive amounts of information.

(2) Reduced Search- Visualizations group information used together, reducing search.- Representation of large amount of data in a small space.

(3) Enhanced Recognition of Patterns- Recognition instead of recall (menu based vs command interface)- Visually organizing data by structural relationships enhances patterns




(4) Perceptual Inference- Visual representations make some problems obvious e.g. geometry of a problem helps to find solutions.- Visualization can enable complex specialized graphical computations.

(5) Perceptual Monitoring- Visualization can allow the monitoring of a large number of potential events if the display is organized so that these stand out by appearance or motion.

(6) Manipulable medium- Unlike static diagrams, visualizations can allow exploration of a space of parameter values and can amplify user operations.




• LifeLineshttp://www.cs.umd.edu/hcil/lifelines/

Lifelines provides a general visualization environment for medical records. It eliminates long lists to scroll, complex searches, endless menus, etc. A one screen overview of the record using timelines provides direct access to the data.

For a patient record, medical problems, hospitalization and medications can be represented as horizontal lines, while icons represent discrete events such as physician consultations, progress notes or tests. Line color and thickness can illustrate relationships or significance. Rescaling tools and filters allow users to focus on part of the information, revealing more details.


http://www.cs.umd.edu/hcil/lifelines/











• Inxight Tree Studiohttp://www.inxight.com/index.html

Inxight Tree Studio allows web site builders to create visual site maps called Star Trees.

The users can see at a glance all the organization of the web site.

Search features that improve the navigation


http://www.inxight.com/index.html




0


• On-line Library of Information Visualization Environments

http://www.otal.umd.edu/Olive/

• List of Information Visualization projects http://www2.iicm.edu/ivis/ivis/node1.htm



1

Mapping Data to Visual Form (1)


Raw Data

Data Tables

Visual Structures

Views Task

Human Interaction

Data

Visual FormData Transformation

Visual Mappings

View Transformations


2


• Raw Data

- Data transformations map raw data, that is, data in some idiosyncratic form, into Data Tables

- Numbers, texts, names, etc -> without structure -> difficult to map to visual forms



3



Raw Data

Data Tables

Visual Structures

Views Task

Human Interaction

Data


Visual Mappings



4


• Data Tables

- This tables are based on relational descriptions of data extended through the use of metadata


Case Case i Case j Case k ...

Variable x Value ix Value jx Value kx ...

Variable y Value iy Value jy Value ky ...

... ... ... ... ...

Metadata are the labels for the rows and columnsThe columns represent cases, sets of values for each of the variables Variables are divided into input and outputs. Input variables uniquely determine the output variables

Casei represents a unique object and the values represent characteristics


5


• Data Tables

- This tables are based on relational descriptions of data extended through the use of metadata.

- Dimensionality is used to refer to the number of input variables, the number of output variables, both together, or even the number of spatial dimension in the data.

- Variable types can be nominal, ordinal or quantitative.

- Errors or missing values and statistical calculations can add additional information. Data tables often contain derived value or structure (e.g. derive the mean [value], sort the variables [structure]).

- Data transformation can help to discovered patterns.



6



Raw Data

Data Tables

Visual Structures

Views Task

Human Interaction

Data


Visual Mappings



7


• Visual Structures

- Visual Mappings transform data tables into visual structures.

- The structures combine spatial substrates, marks and graphical properties.

- A mapping is said to be expressive if all and only the data in the Data Table are also represented in the Visual Structure.

- A mapping is said to be more effective if it is faster to interpret, can convey more distinctions, or leads to fewer errors than some other mapping (e.g. sine wave)



8


• Gestalt laws - principles of organization- There can be interaction among the visual coding of information. - Part of the point of coding information visually is to produce patterns that the eye detects from ensembles of components.


Gestalt principles

• Pragnanz

• Proximity

• Similarity

• Closure• Good continuation

• Common fate

• Familiarity

Every stimulus pattern is seen in such a way that the resulting structureis as simple as possible.

The tendency of objects near one another to be grouped together into a perceptual unit.

If several stimuli are presented together, there is a tendency to see the form in such a way that the similar items are grouped together.

The tendency to unite contours that are very close to each other.

Neighboring elements are grouped together when they are potentially connectedby straight or smoothly curving lines.

Elements that are moving in the same direction seem to be grouped together.

Elements are more likely to form groupsif the groups appear familiar or meaningful.


9



Raw Data

Data Tables

Visual Structures

Views Task

Human Interaction

Data


Visual Mappings



0


• Views

- View Transformations create Views of the Visual Structures by specifying graphical parameters such as position, scaling, and clipping.

- The user can restrict the view to certain data ranges.



1


• Views

- Since spatial position is such a good encoding several techniques have been developed to increase the amount of information that can be encoded with it.

• Composition is the orthogonal placement of axes, creating a 2D metric space. • Alignment is the repetition of an axis at a different position in the space.• Folding is the continuation of an axis in an orthogonal dimension. • Recursion is the repeated subdivision of space.• Overloading is the reuse of the same space for the same Data Table

see-soft



2


• View Transformation

- View transformations interactively modify and augment Visual Structures to turn static presentations into visualization by establishing graphical parameters to create Views of Visual Structures.

Location probes Location proves are view transformations that use location in a visual structure to reveal additional Data Table Information. (filmfinder)

Viewpoint ControlsUse affine transformations to zoom, pan, and clip the viewpoint.



3



Raw Data

Data Tables

Visual Structures

Views Task

Human Interaction

Data


Visual Mappings


186 x 18 --------- 1488 186 _______ 3348

back

Multiplication

back

back

Documents

ICS 280: Information Visualization 1 ICS 280: Advanced Topics in Information Visualization Professor Alfred Kobsa Overview of Information Visualization