También disponible en Español

Inf@Vis!

The digital magazine of InfoVis.net

InfoVis Diagram
by Juan C. Dürsteler [message nº 187]

In this issue we describe the diagram for the process of Information Visualisation as seen by Yuri Engelhardt and the author after a series of discussions about its nature and the process that leads from Data to Understanding.

Following the line of the previous article we propose in the current one a diagram for the process of Information Visualisation that joins some of the elements already seen and adds some new ones. It's important to take into account at least two aspects: 

  • While a restrictive definition of Information Visualisation would place it within the scheme in the phase of the conversion from information into knowledge, where graphics, sound and/or other perceptive stimuli would appear, a wider definition encompasses the whole process of converting data into knowledge whenever a phase including the perception phenomena occurred.

    This way Information Visualisation acts as a vehicle for the building of knowledge, revealing the underlying patterns in data.

  • The definitions of each one of the elements, data, information and even comprehension or understanding are, necessarily, operative definitions, more than being philosophically correct or incorrect. For example, there's an important discussion about the definition of informaction. In this sense our definition as data within context is essentially a simplification since, among other meanings, information can be understood as a complex communication process.

Hence, our definition of Information Visualisation, unlinked to computers or any specific device, and bound to the building of understanding by using perception, leads us to propose a general overview diagram of the Information Visualisation process.

The diagram, product of intense discussions between Yuri Engelhardt (University of Amsterdam) and the author, somewhat combines the four phases of Colin Ware with Shedroff’s conversion from data into knowledge and the more specific view of the transformation that take raw data and convert it into visual mappings present in Card et al.

Esquema_enOK.gif (135015 bytes)
The Infovis diagram.
Source:
Graphic made by the author, based on a previous joint graphic made by Yuri Engelhardt and the author.
Click on the image to enlarge it.

In our proposal, nevertheless, we include explicitly: 

  • the perception and interpretation phase that builds the understanding, transforming representations into perceptions and cognitive artefacts. 

  • the figure of the user and the designer of the visualisation, important players in the process.

  • the encoding process generally performed by the designer that uses a notation schema to encode information that the user has to know or, at least, hast to be able to deduce in order to decode the information and find the pattern he/she is looking for.

The phases of the process 

DatInfo_en.gif (28231 bytes)
Transformation of data into information
Source:
Graphic made by the author.
Click on the image to enlarge it.
  • From Data to Information 
  • Information Visualisation begins with data, that constitutes the building blocks of information. The conversion of data into information has three main phases:: 

    • Collecting and storing data relevant for our study, looking for it in the literature or collecting it directly. 

    • Processing and transforming the data to filter errors, eliminating irrelevant, redundant or spurious data and/or deriving secondary magnitudes from the raw data, for example doing statistics on it.

    • Building data tables that organise the data according to its meaning, converting it in information, data with metadata, data that “explains” other data.


  • Form Information to Visual Representation 

    Once we have the proper information transformed, structured and provided of meaning, the next phase of the process deals with converting this information into a perceptual representation, most of the times in visual form. In this phase we have to identify the best graphical or perceptual structure to represent the information in such a way that the underlying patterns and structures are easy to appear and to be identified.

    InfoRep_en.gif (31736 bytes)
    Transformation of information into visual representation .
    Source:
    Graphic made by the author.
    Click on the image to enlarge it.

    Here we have to use a notational schema, that is a particular visual language that maps information into graphics. In order to be cognitively effective the notational schema has to be somehow in the brain of the user i.e. either the user knows it or can deduce it from previous experience or knowledge.

    Context and culture have to be taken into account for both notational schemas (the one in the designer’s mind and the one in the user’s mind) to match, increasing the possibilities of matching without having to resort to learning new schemas and notations. 

    The designer of the representation can use the elementary objects of graphic representations and attributes like shape, colour, size, transparency among others combined with their location in space and several composition principles to produce a wide range of static or dynamic representations.

    This phase includes all the stages needed to build the representation, like selecting the view, rendering and applying geometrical transformations to represent the data.

  • From Visual Representation to Understanding

    The representation is in place already, but in absence of a human receptor it’s completely useless. In order to help in gaining insight and building knowledge, the visualisation has to produce a perceptual impact on the user.

    RepComp_en.gif (35450 bytes)
    Understanding the visual representation.
    Source:
    Graphic made by the author.
    Click on the image to enlarge it.

    For this to occur the designer has to take into account both sensory representations and arbitrary conventional representations along with some facts of visual perception, cognitive psychology and even linguistics to deploy a visual language with a notational schema that can “enter in resonance” with the notational schemata that the user can understand, be them sensory or arbitrary conventional.

    Modern systems allow users to interact with representations so that they can modify “on the fly” the organisation of data, derive new data or new structures and manipulate the parameters of the representation to select what they want to see and from which perspective. 

    Interaction allows the user to perform physical (enlarging phisically the view) and semantic (presenting more meaning)  zooming, seeing the focus and the context simultaneously(without losing sight of the whole visualisation).

For this reason cultural as well as cognitive aspects have to be taken into account by the designer, that rarely will be the same person as the user. In this point user centred design and usability issues of the representation acquire prominence, since in order to orchestrate a meaningful, easy to understand and interact with representation is by no means an easy task.

Links of this issue:

http://www.infovis.net/printMag.php?num=186&lang=2   Num. 186 about Diagrams for Visualisation
http://www.yuriweb.com   Yuri Engelhardt's personal page
© Copyright InfoVis.net 2000-2016