Conference Paper

Link Proximity Analysis - Clustering Websites by Examining Link Proximity

DOI: 10.1007/978-3-642-15464-5_54 Conference: Research and Advanced Technology for Digital Libraries, 14th European Conference, ECDL 2010, Glasgow, UK, September 6-10, 2010. Proceedings
Source: DBLP


This research-in-progress paper presents a new approach called Link Proximity Analysis (LPA) for identifying related web pages
based on link analysis. In contrast to current techniques, which ignore intra-page link analysis, the one put forth here examines
the relative positioning of links to each other within websites. The approach uses the fact that a clear correlation between
the proximity of links to each other and the subject-relatedness of the linked websites can be observed on nearly every web
page. By statistically analyzing this relationship and measuring the amount of sentences, paragraphs, etc. between two links,
related websites can be automatically, identified as a first study has proven.

Download full-text


Available from: Joeran Beel
This research doesn't cite any other publications.