A new class of business intelligence tools that help reveal information and provide context for it is emerging. These tools provide a visual drill-down capability that can help people understand the interplay of large numbers of data elements and are generally described as being chart-based data exploration tools. So, is there a difference between chart-based data exploration and traditional data visualization? Is it a matter of the old song, "You say to-MAY- to and I say to-MAH-to?" Is the difference superficial, or is there more to it?

One obvious difference is cost. Some might characterize chart-based data exploration as "a poor man's data visualization." Recently, I went on the Web and downloaded a chart-based tool for $39. But the cost of data visualization tools typically starts at $500 per workstation. So there can be a dramatic difference between the two types of tools based solely on price. But what about functionality?

Both data visualization and chart-based data exploration provide the ability for the user to look at data graphically. Cognitive psychologists have long contended that a graphical representation is almost always more comprehensible to a human being than a numerical one. So, in that regard, the two tool types are similar.

Data visualization tools provide elegant functionality to enable human comprehension of data with complex interrelationships. One of the most distinguishing characteristics of data visualization is the ability for the user to interact directly with the data. Among other things, the user can navigate through data, change its scale on the fly, filter it, animate it over time and return to preset viewpoints. Detail numerical data that forms the basis for 3-D analysis can be seen by "brushing," or moving the mouse over a figure.

Chart-based data exploration tools automatically transform tabular data found in spreadsheets, databases and word processing tables into a graphical format (for example, a bar chart). Double-clicking on a single bar within the chart selects subsets of the data. The user can navigate through data by further drill down. With some tools, detail is available within a "statistics window."

With chart- based tools, one-way and two-way tabulations can be viewed. With data visualization tools, three-way tabulations can be viewed using the three spatial dimensions. Color, texture and other features can be set to represent additional dimensions of the data beyond three. Data visualization truly represents multidimensional data.

Setup between the tool types can vary dramatically. With chart-based tools, setup is as easy as copying a block of data from a Windows application to the clipboard and pasting it into the charting application. With data visualization, setup can be more complicated. Data usually has to be in a multidimensional format, which either means it is already housed in a multidimensional database management system or it needs to be moved into some sort of a proprietary multidimensional file system found within the visualization tool. In either case, it takes someone who knows what they are doing to connect a database to the tool.

Data visualization tools can be found in two basic types. One is an ad hoc tool, where several data landscapes or templates have been engineered and provided for use with any data. MineSet from Silicon Graphics and SeeIT from Visible Decisions are examples of tools with predesigned data templates available. The other data visualization tool type is more of an application development environment that can be used to develop landscapes customized for a particular set of data. In3D from Visible Decisions is an example of a data visualization tool that is really an application development environment. Chart-based exploration tools fall into the "template" category; they allow the selection of a limited number of chart types to provide a visual landscape for the data. Examples of products in the chart-based space are Maisy GraphSheet and CrossGraphs.

While there are some fundamental similarities and differences between traditional data visualization and chart-based data exploration, there is perhaps one significant function that differentiates them ­ the number of dimensions that can be visualized at once.

There are times when two-dimensional analysis of data is entirely appropriate. Data that is located in a relational database management system is, by its very nature, two-dimensional ­ comprised of rows and columns. For situations where purely relational data needs to be explored, data visualization might be overkill; whereas chart-based data exploration tools appear to be state-of-the-art in helping people visualize and analyze information. For massive amounts of data or where very complex data interrelationships exist (and multidimensional analysis is assumed), data visualization is the only way to go. The difference is not just a superficial "to-MAY-to" versus "to-MAH-to" issue. Either way, we have tools that can really make a difference in our ability to understand and act on information.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access