2008-10-14: WET: The Website Exploration Tool demo still available at www.infovis.net

From InfoVis:Wiki
Revision as of 16:43, 14 October 2008 by Vpascual (talk | contribs) (New page: Category:NewsCategory:2008/10 http://www.dtic.upf.edu/~vpascual/WET.html The Website Exploration Tool (WET) is a visual tool to support Web Mining. Its main characteristic is its...)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search


http://www.dtic.upf.edu/~vpascual/WET.html

The Website Exploration Tool (WET) is a visual tool to support Web Mining. Its main characteristic is its ability to combine the visualisation of data based on the structure, content and usage of a website.

WET automatically extracts descriptive hierarchies of a website enabling the visualisation of its structure and users' navigation along with its contents and usage. To achieve this, a set of web metrics (e.g. number of hits per page, page rank or landing pages, among others) can be mapped into the visual attributes of glyphs that form a representative hierarchical visual metaphor of the site. In addition, all the available visualisations may be combined through a linking and brushing system that enhances the multi-metaphor user experience.


The architecture of WET


The Information Retrieval and Web Mining System (IRWMS): This system uses a crawler to get the structure and textual content of the target website, and calculates a set of web metrics from the usage data collected from the webserver log files and a logging javascript system. This data is finally stored into a database.

The Graph's Logic System (GLS): Representing a very large graph on a display area is a very complicate task, often not as useful as simplifying the amount of information to be shown. Hence, the GLS is capable of extracting representative subgraphs from the whole collected webgraph, simplifying the quantity of data to be visualised.

The Visualisation System (VS): A GraphML file is generated from the output of the IRWM and GLS systems, and used as input to the VS, which depicts the information and enables the analysis and exploration of the data.


WET is currently available as a demo at http://www.infovis.net/printRec.php?rec=revista&lang=2, and will be fully available for test purposes very soon.