Classifcation and comparison of information structures from a web page
Mirel Coşulschi, Nicolae Constantinescu, Mihai Gabroveanu · Annals of the University of Craiova Mathematics and Computer Science Series · 2004
This paper is closely related to the web structure analysis based on visual representation direction, a new and promising one, from which many web application such as information retrieval, information extraction can take advantage of. Page segmentation can be used to improve the query process in information retrieval, to simplify the work of automatic wrappers and to eliminate irrelevant data such as navigational bar and advertisement. Simple DOM based segmentation did not show satisfactory results, but the combination of DOM features with visual characteristics lead to a better partitioning. 2000 Mathematics Subject Classi cation. 68P20,68U99.