WVFA: a Web visualization framework architecture

Nowadays, humans and machines don't use the same format of knowledge representation. This gap represents a serious problem in the interaction human-machine. More specifically, comprehension of text and images plays a key role in the process underlying Web navigation. This fact aim us to explode the application of intelligent methods that minimize the gap between the information that is showed to users and that they are able to understand and use. This paper presents our architecture proposal to improve Web usability: firstly, providing a procedure to enable conceptual modeling of Web interfaces at high level that is close to human visual perception. Then, moving this modeling to a perceptual layer with gradually decreasing the level of abstraction. This step will bridge the gap between the human use and the computation-centered mechanisms, combining data and treatment exhibition in a per-site and per-user basis.