Short Abstract
Ranking is essential in information retrieval systems, as the web serves as a vast repository of both static and dynamic pages, providing an infinite source of information enriched with numerous hyperlinks. This wealth of data necessitates effective ranking to meet user needs, as search engines encompass a significant portion of web pages. The PageRank algorithm measures the importance and behavior of web pages, enhancing the relevance of search results through analysis of the web’s graph structure. In contrast, the Hypertext Induced Topic Search (HITS) algorithm operates on a framework of hubs and authorities, leveraging the rich context of hyperlinks. This paper analyzes web page importance using various ranking algorithms and introduces a novel algorithm, LinkRanker, to enhance ranking efficiency. Its performance is compared against established algorithms, including HITS, the Stochastic Approach for Link-Structure Analysis (SALSA), and Norm(p), with evaluations conducted on multiple hyperlink graphs.
2026, Journal
https://link.springer.com/article/10.1007/s40009-026-01958-6