WSOLINK: web structure outlier detection algorithm

Abstract

In this world of specialisation where everything is getting specialised, data warehouses and web mining techniques are also getting specialised. Web usage mining, web content mining, and web structure mining are various categories of web mining techniques depending upon the data to be mined. Apriori algorithm, FP growth algorithm, and average linear time algorithm are available to analyse the general access patterns in web server logs whereas WCOND-mine and signed with weight technique are web content outlier mining algorithms. However, no such algorithm is available to check the authenticity and availability of hyperlinks in the resultant web pages given by web search engines. The present research work aims at detection of outliers from the results of queries over web pages through web search engines.