Graphics Reference
In-Depth Information
Figure
.
.
A tree view clustering representation
Interactivity Within and Between Graphics
4.7.2
Global views are used to provide context for more detailed views, to help formulate
a search, identify patterns, or make a gestalt overview (Cleveland,
,
; Boun-
ford,
).
Human-machineinteractionconferssubstantialadvantagestothepowerofstatis-
ticalgraphics(ClevelandandMcGill,
).Interactivemanipulationisthecapability
to directly interact with displays to query them for information, to link them, and to
changetheirshapes,scales,andothercharacteristicssothat awiderangeofviewscan
be scanned. Interactive manipulation techniques can be developed and used within
each graphic as well as between graphics and between graphics and data.
Manipulation techniques can be classified into two main areas:
. Interactivity within graphics, aimed at promoting interactivity on a single
graphical representation. Among the available tools are: the ability to show/hide
parts of a graph, the ability to use color/size information in order to change the
representation according tothe point attributes, and the ability to usethe mouse
pointer to query the graphical representation. Moreover, the user can alter the
data table in order to observe changes on the graph. Such changes include the
deletion of data, the highlighting of data (the data will stand out against normal
data), and the focus of data (in order to show as mush detail as possible on the
data).Nevertheless,traditionaltoolscanbeused,i.e.,context-sensitivequerying,
zooming, flexible scaling, resizing of graphics andof objects, coloring of objects,
selecting (by points, areas, etc.), masking, and grouping (Dhillon et al.,
;
Hurleyand Buja,
).
. Interactivitybetweengraphics,aimedatpromotingtheinteractivityamong
different views of the same data. It is straightforward to say that the previously
listed tools can be used also for this kind of interactivity, mainly based on the
linking concept: to any action on a view corresponds the same reaction on all
the related views of the same data (Symanzik,
).
