|InfoVis.net>Magazine>message nº 66||Published 2001-11-12|
|También disponible en Español|
The digital magazine of InfoVis.net
In order to do this we'll follow the scheme that Stephen G. Eick of Visual Insights depicts in his suggestive article "Visualizing Online Activity" (Communications of the ACM, August 2001; available at ACM digital library under subscription).
According to Eick, there are three areas where information visualisation can help us to understand our visitor's activity:
We already spoke about section 1 in the past issues 47 and 48 of the e-zine but it's still worth remembering the existence of systems based on hyperbolic geometry to represent the structure of large web sites. Among them those provided by Inxight or Silicon Graphics Site Manager that uses the technology developed by Tamara Munzner as part of her PhD thesis.
So we'll concentrate in this issue on the visualisation of visitor's flow and individual paths. The goal of this type of tool is the analysis of the entry points of our visitors, what pages they visit and where they leave the web. Path analysis concentrates on what individual users do within the web site, while flow analysis integrates the paths and statistics of many users in order to reveal the global usage patterns of the site as a whole.
Examples of path analysis can be found at Advanced Log Analyzer from Vasin Knoware. This analyser allows you to visualise the individual paths of the last 30 visitors or an aggregate of the most frequent paths among many other statistics. In both cases the visualisation is made in a tabular style.
Sawmill offers a visualisation of individual paths à la Windows Explorer that is quite clear and occupies less space than the previous case.
A much more attractive visual system is the one that VISVIP providesl, developed by John Cugini as part of the interesting set of tools WebMetrics from the NIST. VISVIP represents the structure of the web site as a 2D graph over which the path of a particular user is superimposed as a smoothly curved line that jumps from page to page. Several arrows included in the line show the direction and dotted lines proportional in length to the time spent are placed in each visited page.
As an example of flow analysers, Eick himself proposes the graphical tools of their own solution called Ebizinsights, software devoted to the e-business sites that represents data in several graphical forms (see the Gallery at Visual Insights site).
In this case the flow between pages is represented by a rectangle (the page in focus) with lines that link it to the pages where the users have come from and with other pages where the users have left. Each page contains the percentage of visitors coming in or going out. Clicking on a particular page, it becomes the centre and shows all the pages where users come and go.
Next week we'll continue exploring visualisations of visitor behaviour within a web site.
Links of this issue:
Subscribe to the free newsletter