KernelRank: Exploiting Semantic Linkage Kernelsfor Relevant Pages Finding

Relevant pages finding is to find a set of relevantpages that address the same topic as the given page. Hyperlinkrelationship is an important useful clue for this task. Some hyperlinksare useful, also some are irrelevant or noisy. Therefore, it isimportant to design efficient relevant pages finding methods that canwork well in the real-world Web data. In this paper, we propose arelevant pages finding algorithm, KernelRank. This algorithm takesadvantage of linkage kernels to reveal latent semantic relationshipsamong pages and to identify relevant pages precisely and effectively.Experiments are conducted on WT10G and the results show that theKernelRank algorithm is feasible and effective.