Categorize web sites based on design issues

Abstract

Interface design is one of the most important topics during web development process. The final design is a tradeoff between the owner's personal idea and the web developer's perception of what he wants. In this paper, we have proposed a new model called WLDM (Web layout design model) to cover the important components of interface design. There are three components in the WLDM, including structure, content and visual. We have selected three features for structure, two for content, and three for visual component. Thereafter, we have made a dataset using 1088 most visited web sites. Finally, applying K-means algorithm, we have clustered this dataset. According to our result, six clusters were identified. Considering WLDM, web layout designer have a blueprint to cover areas of research related to this issue. The result of this clustering can be used for recommender systems to map owner groups, which have different attitude. © 2011 Springer-Verlag.