Theme 4: Mapping Process

This theme discusses a central part of the visualization process, namely the mapping of data to a representation that can be (visually) perceived. Different strategies of the mapping process (such as by Renaissance teams, top-down or bottom-up methods) are to be explained. Possible difficulties, such as artifacts, are emphasized. Design factors, such as annotations or scale bars, are necessary to provide visual context.

(4.1) Models of mapping strategies

This chapter explains various approaches used to bind images to data.

(4.1.1) Blind matching procedure

The "blind matching procedure" demands a preprocessing stage of data and a set of visual attributes and graphics primitives. First the data is separated into individual data elements that are subsequently matched to visual elements (graphics primitives and visual attributes) to build up one or more images. Matching between data elements and visual elements are continued until the user is satisfied with the results.

(4.1.2) Renaissance teams

Renaissance teams bring together experts from various disciplines, such as from computer graphics, art and design, and the sciences. Matching between data elements and visual elements is done methodically utilizing the expertise of team members.

(4.1.3) Top down versus bottom up strategies

Bottom up strategies match data elements to visual elements in order to create one or more images. Such generated images need to be tested for their value to the user (e.g. coherency, expressiveness, effectiveness). Top down strategies start out with coherent, easy to interpret images and match their visual elements to data elements.

(4.1.4) Generate-and-test

Generate-and-test is a procedure to test images (after the bottom-up approach) for their value to the user. This test can be done automatically.

(4.1.5) Case based reasoning

Case based reasoning is a procedure mostly used with top-down matching strategies. Case based reasoning adapts and combines successful visual representations by keeping a data base of such representations for use and manipulation by a user.

(4.2) Difficulties in the mapping process

Computer-generated representations can be difficult to interpret due to e.g. approximations or artifacts.

(4.2.1) Approximations

introduced by the representation technique, as when a smooth surface is approximated by a collection of polygons

(4.2.2) Artifacts

introduced by computer graphics techniques, as when a surface color varies because of the lighting model

(4.3) Visual context

Design Issues, such as scale bars or annotations provide additional visual context outside the main mapping stage.