ON TWO ALGORITHMS USED IN WEB STRUCTURE MINING

Due to the continuous growth and spread of the internet using Web Mining to improve the quality of different services has become a necessity. Web Mining is nothing else than applying data mining techniques and algorithms on web data. In this work we present two algorithms used in Web Structure Mining namely Page Rank and HITS. Both algorithms draw their origin from social networks analysis and they are modeled based on the Theory of Markov Chains. Page Rank is used by the search engine GOOGLE and HITS by the search engine CLEVER. We present their strengths, weakness and other areas of applicability.