Visualization Process

There are many other visualization techniques and systems shown in the tables. We need to
organise methods into categories or taxonomies to structure the study of field.
• Most visualization pipelines and systems map easily to these stages and
transformation or computation can be placed at any of these stages.
• Two key points are:
1. User Interaction can take place at any point in this pipeline(nodes & links)
2. Each link is a many-to-many mapping.
• Steps that need to be taken to define a visualization of data, transformations
and processes that alter the data.
1. Data Pre-processing and Transformation.
2. Mapping for Visualization.
3. Rendering Transformations.
 Measures of Visualizations
• Measures that define mathematically and modifications can be applied at all
stages of the pipeline.
❑Expressiveness
❑Effectiveness
• An Expressive visualization presents all the information and only the
information, that measures concentration of information.
• We measure it as Mexp of that information,divided by the information we
want to present it to the user.
0<=Mexp<=1
➢If Mexp = 1, Ideal Expressiveness.
➢If Mexp<1, the information displayed is less than that desired to be
presented.
➢If Mexp>1, presenting too much information.
• Effectiveness:
A visualization is effective when it can be interpreted accurately &
quickly and when it can be rendered in a cost-effective manner.
• Effectiveness thus measures a specific cost of information perception.
• If the interpretation time increases,either due to the increasing complexity or
the size of the dataset, Meff decreases, emphasizing the rendering time.
• We define
➢Meff = 1/(1+interpret+render)
➢0<Meff<=1
• The larger the Meff is, the greater the visualization’s effectiveness.
• If Meff is small,then either interpretation time is very large or the rendering
time is large.
• If Meff is large, then both the interpretation and the rendering time are very
small.
 Semiology of Graphical symbols
• Every visual object is called a graphical symbol, which are easily recognised.
The science of graphical symbols and marks is called semiology.

• Every possible construction in the Euclidean plane is a graphical representation

made up of graphical symbols.

• Semiology uses the qualities of the plane and objects on the plane to produce
similarity features,ordering features, and proportionality features of the data
that are visible for human consumption.

• There are numerous characteristics of visualizations, of images, or of graphics

made up of symbols. Some of these are:
➢Symbols and Visualizations
➢Features of Graphics
 Symbols and Visualizations
• Contains an image that is universally recognizable.(yield sign), such images
preattentively recognizable with experience.
• Contains an image which requires a great deal of attention to understand.
• Fig(1) is perceived in one step, which is simply an association of its
meaning.
• Fig(2) needs two steps for understanding, first identifies the major
elements of the image, second identifying the various relationships
between them.
• Steps include patterns, subset of groups,information having perceptual or
cognitive commonality, last step is to identify the clusters
• Without external identification(cognitive),a graphic is unusable
• Discovery of relations or patterns occurs through two main steps, the first is
the mapping between any relationship of the graphic symbols and the data
that these symbols represent.
• Any pattern on the screen must imply a pattern in the data.
• Any perceived order in graphic symbols is directly corelated with a perceived
corresponding order between the data, and viceversa.
• If some order pops out visually and such order is not present in the data, we
are misled.
• So,
❑Similarity in data structure  Visual similarity of corresponding symbols.
❑Order between data items  Visual ordering between corresponding symbols
 Features of Graphics
• Graphics have three are more dimensions
• Shows a matrix with points of various sizes within each
cell.
• Every point in the graphic can be interpreted as a relation
between a position in “X” and a position in “y”.
• Set of all points either in the 2D or 3D image represents
the totality of the relations among the three dimensions
x,y,z and any patterns present imply a pattern in the data.
 Features of Graphics
• When looking at this figure we immediately see two
branches.
• The eye sees either branch independent from the
number of its leaves.
• Graphic can contain a large number of similar
items,themselves graphics, with the only possible
limitations being technical ones, such as that the various
graphic symbols are distinguishable from each another.
• The eye can preatentively see the various distribution of
symbols.
 Rules of a Graphic
• All objects will be interpreted as flat(2D) or as physical objects (in 3D). Some of the
fundamental rules are:
➢ The aim of the graphic is to discover groups or orders in x, and groups or orders in y, that
are formed on Z-values.
➢ (x,y,z) – construction enables in all the cases the discovery of these groups.
➢ Within the (x,y,z)- construction ,permuatations and classifications solve the problem of the
upper level of information.
➢ Every graphic with more than three factors that differs from the (x,y,z) – construction
destroys the unity of the graphic and the upper level of information.
➢ Pictures must be read and understood by the user.
 The Eight Visual Variables
• Application of graphics to communicate information requires an
understanding of graphic primitives and their properties.
• Graphic primitives are termed as Marks.
• The positioning of marks within a display space provides a mean to map or
reveal additional properties of the underlying data, including similarity and
distributions.
• Marks can vary in size, can be displayed using different colors, and can be
mapped to different orientations, all of which can be driven by data to convey
information.
• There are Eight ways in which graphical objects can encode information, and
these eight variables can adjusted as necessary to maximise the effectiveness
of a visualizationto convey information.
 EIGHT VARIABLES
❑ Position.
❑ Mark.
❑ Size(Length, Area and Volume)
❑ Brightness
❑ Color
❑ Orientation.
❑ Texture.
❑ Motion.
 POSITION
• The placement of representative graphics within some display space be it
one,two,three dimensional.
• Position has the greatest impact on the display of information, because the
spatial arrangement of graphics is the first step in reading the visualization.
• The best positioning scheme maps each graphic to unique positions, such that
all the graphics can be seen with no overlaps.
• The selection of variables used to organise data within a display can answer
the variety of questions such as:
➢Where do most of the data values fall ?
➢Does the data fit any well-known statistical distribution ?
➢Are there visible trends in the data ?
➢Are there any clusters and structures within the data ?

• In selecting appropriate variables to organise the display space and to

present values with representative graphics, three features are used,
1. Scale.
2. Projection.
3. Axes
• Scale:
Scales can be applied to variables to remap values to reveal
structures. Two types of scales ,
➢Linear scale – mapping is simply an offset, and a stretching or shrinking of
the range.
➢Logarithmic scale – used to map exponentially increasing variables into
more compact ranges.
• Projection:
The mapping of two or more variables more correctly termed a
projection from one data space to another., the second generally of a lower
dimensionality.
• Axes:
These are graphical elements that provide additional information for
understanding how the visual space is defined. It contains tick-marks
indicating intervals of the data range, and text labels that provide data
values.
 MARK
• It is defined as a shape:points,lines,areas,volumes, and their
compositions.
• Any graphical object can be used as a mark,including symbols, letters and
words.
• When using marks ,it is important to consider how well one mark can be
differentiated from another marks.
• Within a visualization, there are hundreds or thousands of marks to
observe, we try not to select marks that are too similar.
 Size(Length,Area and Volume)
• Position & marks are used to define visualization, remaining variables
affect the way individual representations are displayed.
• Size determines how small or large a mark will be drawn.
• Size easily maps to interval and continuous data variables, because that
property supports gradual increments over some range.
 BRIGHTNESS
• It is also called as luminance is the variable used to modify marks to
encode additional data variables.
• Brightness can be used to provide relative difference for large interval &
continuous data variables, or for accurate marks distinction for marks
drawn using a reduced sample brightness scale.
• Visualization using brightness display information
 COLOR
• Color can be defined by two parameters,
➢Hue:provides what most think as a color – dominant wavelength from
the visual spectrum.
➢Saturation: level of hue relative to gray, and drives the purity of the color
to be displayed.
• Use of color to display information requires mapping data values to
individual colors.
• Color maps are useful for bothinterval and continuous range of hue and
saturation values
 ORIENTATION
• This graphic property describes how a mark is rotated in connection with
a data variable.
• The best marks used for orientation are those with a natural single axis,
the graphic exhibits symmetry about a major axis.
• It displays a mark that looks like a elongated triangle, which clearly
defines a single major axis.
 TEXTURE
• Texture can be combined with a combination of many of the other visual
variables,including marks(texture elements),color(associated with each
pixel in a texture region) and orientation(conveyed by changes in the
local color).
• Varying the color of the segments or dots can also be perceived as a
texture.
• Texture is most commontly associated with a polygon,region or surface.
• In 3D, a texture can be an attribute of a geometry , such as of ridges with
varying height frequency and orientation.
• The distribution and orientation of marks themselves can form regions of
texture.
 MOTION
• Motion can be associated with any of the other visual variables , since
the way a variable changes over time can convey more information.
• The other aspect of motion is in the direction; for position, this can be
up,down,left,right,diagonal,or basically any slope, while for other
variables it can be larger/smaller, brighter/dimmer, steeper/shallower
angles and so on.
 Bertin’s Semiology of Graphics
• A science dealing with sign systems.
• Keypoint is strictly separation of content(the information to encode) from
the container(the properties of the graphic system).
• A graphics is defined as a combination of a plane properties (implantation
plus imposition) and retinal variables(visual variables above the plane)
• Plan for bertin is a plain sheet of white paper of a standard size, “visible at
a glance”.
• Implantation: Figures of plane geometry, points,lines and areas.
• Impositions: Two planar dimensions dividing into diagrams,netwoks,maps
and symbols.
• Retinal variables ,graphic variations designed for visual perception.
• There are six retinal variables identified by experimental psychology:
➢Size(height,are or number)
➢Value(saturation)
➢Texture()fineness or coarseness)
➢Color(hue)
➢Orientation(angular displacement)
➢Shape

Marks Points,lines and areas

Positional Two planar dimensions
Retinal Size,value,texture,color,orientation and shape

Bertin’s Graphical vocabulary

 Mackinlay(1986) APT
• A Presentation Tool was design to extract some information from the
database and to render a graphical image that presented this information.
• Mackinlay went on to describe graphical languages.
• The syntax of a graphical language is defined to be a set of graphical
sentences.
• Two graphic design criteria were then identified for these languages.
➢Expressiveness – stating that the graphical languages must express the
desired information.
➢Effectiveness – requiring for any given situation that the graphical
languages must effectively utilize the display medium and human visual
system.
• Mackinlay’s work pertains to his composition algebra, a collection of
primitive graphic languages and composition operators that can form
complex presentations
Marks Points,lines and areas
Positonal 1D,2D and 3D
Temporal Animation
Retinal Color,shape,size,orientation,saturation and texture

Mackinlay’s Graphical vocabulary

Encoding Technique Primitive Graphical Language

Retinal-list Color,shape,size,saturation,orientation,texture
Single-position Horizantal axis,vertical axis
Apposed-position Line char,Bar chart,Plot chart
Map Road map,Topographic map
Connection Tree,Acyclic graph,network
Misc(angle,contain,…….) Pie chart,Venn diagram

Mackinlay’s basis set of primitive graphical languages

 Bergeron and Grinstein Visualization reference model
• Describes an Abstraction of the visualization problem, which is mapping from
the underlying data space to a physical representation.
• Model is represented by Visualization pipeline. It is defined as four stages.
➢Fist stage: Identifies source and provides appropriate information about the
data structure.
➢Second stage: Standardised data enters the model transformation change
which defines appropriate projections of the source data space to a usable
representation data space
➢View specification stage identifies the appropriate mapping from the
transformed data space to the visual representations.
➢Association stage performs the generations of graphics defined by the
representations and end encoded with the data ,resulting in the perceptual
simulation of the data,including both graphic and sound representations
 Wehrend and Lewis
• They defined a mechanism that constructs a large catalog which is
arranged as a two-dimensional matrix classified as objects and
operations.
• The objects identify the problems and are grouped together based on
their target domains, while the operations identify groups of similar
goals.
• Catalog is filled with problems(tasks to perform) and
solutions(visualization techniques that provide answers).
• This catalog would prove especially useful as a visualization knowledge
base for such a system.
 Robertson natural scene paradigm
• It aims to visually display data represented by identifiable properties of
realistic scenes.
• Natural scene views are defined as two or three dimensional spatial
surfaces with spectral and temporal variables.
• Visual properties such as surface height,material,density,pase and
wetness are defined and ranked, based on perceptual characteristics.
 Casner(1991) BOZ
• BOZ is an automated graphic design and presentation tool used to replace logical task descriptionjs with
perceptually equivalent tasks by encoding
• Logical interferences(mental arithmetic or numerical comparisons) with
• Perceptual interferences(shortest distance and average size) from which solutions can be visually obtained.

• BOZ designs a graphic with two aims:

1. making logical facts more clear to the user
2. help searching for needed information

• BOZ´s task-analytic approach uses the following five components:

1. a logical task description language
2. a perceptual operator substitution
3. perceptual data structuring
4. a perceptual operator selection
5. a rendering component
 Beshers and Feiner Autovisual
• AutoVisual is an automatic system for designing visualizations within the n-
vision which implements the worlds-within-worlds visualization technique that
recursively defines Sub-space coordinate system.

• It is defined as the hierarchy of interactors consisting of four components:

1)Encoding objects,
2)Encoding spaces,
3)Selections and a
4)User Interface.

• It supports interactive visualizations which use a modifies version of

Mackinlay’s expressiveness & effectiveness criteria.
 Senay and Ignitius VISTA
• They focused on Scientific data visualization and developed VISTA(visualization
tool Assistant),a knowledge based system for visualization design.
• It implements composition rules for the generation of complex visualizations
from simple visualization techniques.
• They start by identifying three sub-processes defining the visualization
pipeline.
Data manipulation, Visualization mapping and Rendering.
• VISTA’s marks encode data using three variation methods.
➢Positional
➢Temporal
➢Retinal
 VISTA’s Composition Rules
• Mark Composition:
Marks with a compatible set of marks.

• Composition by super Imposition:

Superimposing one mark set to another.

• Composition by union:
Combines marks using set union.

• Composition by transparency:
Manipulating the opacity values of marks.

• Composition by Intersection:
Computes the Intersection of visualization
techniques.
 Golovchinsky AVE
• AVE(Automatic Visualization Environment) is an automatic graphical presentation systems
based on generative theory or diagram design.

• Diagrams are composed of graphical elements – only rectangles as implementation.

• The resulting graphics are trees and graphs depicting nodes as rectangles and relationships
with lines or arrows.

• AVE consists of four stages:

➢ Query – Database views such as tuples, domains, Frames.
➢ Analysis – Groups tuples based on relationship type & Frame type.
➢ Constraint network generation – takes slot-bases & type-based relation and maps them to
one or more graphical relations.
➢ Geometric layout – calculated appropriate object positions
 Card, Mackinlay & Shneiderman Spatial Substrate
• They presented a Reference model for visualizations describing three primary
transformations – Data transformations, Visual mapping and View
transformations.

• Spatial Substrate deals with the use of spatial positioning for encoding data
within the display, in this the structure is described as Axes and their
properties and presented four elementary types of Axes.

1. Unstructured(No axis)
2. Nominal(a region is divided into sub-regions)
3. Ordinal(the ordering of these sub-regions is meaningful),
4. Quantitative(a region has a metric, possibly an interval or ratio or specialized)
• Five techniques are described for the increasing the amount of information
that can be encoded by spatial positions.

➢Composition of Axes – orthogonal placement of axes creating a 2-D metric

space.
➢Alignment of Axes – repetition of axes at a different position in space.
➢Application of recursion to axes – repeated subdivision of space.
➢Folding of Axes – continuation of an axis in an orthogonal direction.
➢Overloading of Axes – resuing the same display space.
 Kamps EAVE
• EAVE (Extended Automatic Visualization Engine) by kamps is an extension of
AVE.

• EAVE takes arbitrary relations as input and generates diagram visualizations.

• Automatic design of diagrams is accomplished through three-phase process:

➢Classification phase – properties of data.

➢Graphical resource allocation phase – encode data relations using graphical
relations.
➢Layout phase – realization of design decisions.
 Wilkinson Grammar of Graphics
• It is actually a grammar of statistical visualizations ,a subclass of data
visualizations readily used for statistical analysis..
• There are three primary stages for graphic creation:
➢Specification – Identifies the numerous object transformations &
representational mappings that define the display.
➢Assembly & Display are important for the final visual output of the graphic.
• Grammar specification is composed of seven parts, data and trans operate on
the data space.
• Frame, scale & coord define the geometry of the graphic and arrangement of
data variables.
• Graph defines visual representations, guide defines individual labels and
markers.
List of Wilkinson’s specifications and aesthetic attributes.
Data Set of data operations that create variables from data sets
Trans Data variable transformations
Frame A set of variables, related by operations, that define a space
Scale Scale transformation
Coord Coordinate system
Graph Graph and their aesthetic attributes
Guide One or more guides for interpreting a graph

Form Surface Motion sound Text

Position Color Direction Time Label
stack hue Speed Volume
dodge saturation Acceleration Rhythm
jitter brightness Voice
Size Texture
Shape pattern
polygon granularity
glyph orientation
image Blur
Rotation Transparency
 Hoffman Table Visualizations
• It is a first attempt at defining a generalized space of data visualizations.

• This table model came into existence by four research specific visualizations
techniques called survey plots, scatter plots, RadViz and parallel coordinates.

• Table visualizations are defined as graphic presentations for two dimensional

data tables.

• Primitives are dimensional anchor, a graphic curve defining a mathematical

axis associated with a particular data dimension.
Hoffman’s dimensional anchor graphical parameters
 Taxonomies

• A Taxonomy is a means to convey a Classification often hierarchical in nature,

used to group similar objects and define relationships.
• In visualization, there are many forms of taxonomies, including data,
visualization techniques, tasks and methods for interaction.

➢Keller & Keller taxonomy of visualization goals.

➢Shneiderman data type by task taxonomy.
➢Keim Information visualization Classification
 Keller and Keller Taxonomy of visualization goals
• He classified visualization techniques based on the type of data being analysed and the
user’s tasks.
• Data types considered are:
1. Scalar.(or scalar field)
2. Nominal
3. Direction(or direction field)
4. Shape
5. Position
6. Spatially extended region or Object(SERO).
• Task list consists of :
➢ Identify – establish characteristics by which an object is recognizable.
➢ Locate – ascertain the position(absolute or relative)
➢ Distinguish – recognize as distinct or different.
➢Categorize – place into classes or divisions.
➢Cluster – group similar objects.
➢Rank – assign an order or positive relative to other objects.
➢Compare – notice similarities & differences.
➢Associate - link or join the relationship that may or may not be of the
same type.
➢Corelate – establish a direct connection such as casual or reciprocal
 Shneirderman Data type by Task Taxonomy
• His data types included more types from the information visualization field.
List of data types consisted of :
One-dimensional linear, Two-dimensional map, three-dimensional world,
temporal, multidimensional, tree, network.
• Task set consisted of more extracting knowledge from the data, such as
➢Overview : Gain an overview of the entire collection,ex: using a fisheye
strategy for network browsing.
➢Zoom: Zoom in items of ointerest to gain a more detailed view, ex: holding
down a mouse button to enlarge the region of the display.
➢Filter: Filter out uninteresting items to allow the user to reduces the size of a
search ex: dynamic queries that can be invoked via sliders.
➢Detail-on-demand: Select an item or group and get details when needed.ex: a
pop-up window that shows details of a specific object.
➢Relate: View relationships among items. Ex: select a particular object that
can then show all other objects related to it.
➢History: Keep a history to allow undo, replay and progressive refinement.
➢Extract: Extract the items or data in a format that would facilitate other
uses i.e., saving to file, sending via e-mail, printing.
 Keim Information Visualization Classification
• Keim designed a classification scheme for visualization techniques, and
interactive/distortion methods. The components of each of his classification
dimensions are:
• Classification of data types: 6 types of data exist:
1. One-dimensional data: temporal data, news data, stock prices, text
documents.
2. Two-dimensional data: maps, charts, floor plans, newspaper lauouts.
3. Multidimensional data: Spreadsheets, relational tables.
4. Text and Hypertext: new articles & web documents.
5. Hierarchies and graphs: telephone/network traffic, system dymanics models.
6. Algorithm and software: software, execution traces, memory dumps
 Classification of Interaction & Distortion
Classification of Visualization Techniques Techniques

• Standard 2D/3D displays – x,y or x,y,z – • Dynamic projection – grand tour system,
plots, bar charts, line graphs. Xgobi
• Geometrically transformed displays – • Interactive filtering – Magic lenses,
landscapes, scatterplots matrices, infocrystal, dynamic queries, polaris
prosection views, hyperslice, parallel • Interactive Zooming - Tablelens, Dataspace,
coordinates MGV and scalable framework.
• Iconic displays – needle icons, star icons, • Interactive distortion – hyperbolic &
color icons, tile bars. spherical distortions, bifocal displays,
• Dense pixel displays – recursive patterns, graphical fisheye views.
circle segments, graph sketches. • Interactive linking & brushing – Multiple
• Stacked –displays – hierarchical axes, scatterplots, barcharts, parallel coordinates,
worlds-within-worlds, tree maps, cone trees pixel displays & maps.
Classification of Information Visualization Techniques
 Gibson’s Affordance theory
• According to Gibson, affordances are physical properties of the Environment that we
directly perceive.
• We think perception as a very active process, brain deduces certain things about the
environment based on the available senses.
• Gibson concentrated on the visual system as a whole and not to break perceptual
processing into components n & operations.
• He described term “Resonating” to describe the way the visual system responds to
properties of the environment.
• There are three problems with Gibson’s direct perception approach in developing theories
of how visualizations work -
• First problem – Even if perception of the environment is direct , it is clear that
visualization of data through computer graphics is very indirect.
Many layers between data and its representation.
Source of the data may be microscopic
Source of the data may be abstract.

• Second problem – There are no clear physical affordances in any graphical user
interface.
Use of buttons is arbitrary, when pressed they do interesting things in the real
world.
Perception and action are linked in even more indirect ways when we use a
computer – learn that a picture of a button can be “pressed” using a mouse,
cursor or any other button.
• Third problem – Gibson’s rejection of visual mechanism is a problem.
Color is based on years of experimentation, analysis, and modelling of the
perceptual mechanisms. Color television and many other display technologies
are based on an understanding of these mechanisms.
Understanding of perceptual mechanisms is basic to providing visualization
designers with sound design principles.
 A Model of Perceptual Processing
• A simplified information processing model of human visual perception.
• It gives a broad schematic overview of a three-stage model of
perception.
• In stage 1, information is processed in parallel to extract basic features of
the environment.
• In stage 2, active processes of pattern perception pull out structures and
segment the visual scene into regions of different color, texture, and
motion patterns.
• In stage 3, the information is reduced to only a few objects held in visual
working memory to form the basis of visual thinking.
 Stage 1. Parallel processing to extract low-level properties
• Visual information is first processed by large neurons in the eye and in the primary visual
cortex at the back of the brain.
• Individual neurons are selectively tuned to certain kinds of information, such as orientation
of edges or the color of a patch of light.
• In stage 1 processing, billions of neurons work in parallel, extracting features from every part
of the visual field simultaneously.
• Important characteristics of stage 1 processing include:
➢ Rapid parallel processing
➢ Extraction of features, orientation, color, texture, and movement patterns.
➢ Transitory nature of information, which is briefly held in an iconic store.
➢ Bottom-up, data-driven model of processing
➢ Serving as the basis for understanding the visual salience of elements in displays
 Stage 2. Pattern perception
• At the second stage of visual analysis, rapid active processes divide the visual
field into regions and simple patterns, such as continuous contours, regions of
the same color, and regions of the same texture.
• The pattern-finding stage of visual processing is extremely flexible, influenced
both by the massive amount of information available from stage 1 parallel
processing and b the top-down action of attention driven by visual queries.
• Important characteristic of stage 2 processing include:
• Slower serial processing
• Top-down attention being critical to the formation of objects and pattrns pulled
out from the feature maps.
• A small number patterns becoming bound and held for a second or two under
top-down attentional processes.
• Different pathways from object recognition and visually guided hand motion(the
perception and the action channels).
• There is a major fork in the pattern-processing pathway, with one branch
leading to object perception and the other branch leading to parts of the brain
involved in the control of actions.
• This is the basis for the two-visual-system theory:
➢One system for locomotion & action called action system
➢Another for object identification called the What system
 Stage 3. Visual Working memory
• This is the highest level of perception , where the objects held in visual working
memory by the demands of active attention.
• For External visualization, we construct a sequence of visual queries that are
answered through visual search strategies.
• At this level, only a few objects can be held at a time,vconstructed from
available patterns that may provide answer to the visual query and from
information stored in long-term memory related to the task.
• For example , the road map to look for a route, the visual query will trigger a
search for connected red contours(representing major highways) between two
visual symbols(representing cities)