Web-Page Clustering Using Cemetery Organization Behavior of Ants

Abstract

Clustering is the unsupervised classification of patterns (observations, data items, or fea­ture vectors) into groups (clusters). The clustering problem has been addressed in many contexts and by researchers in many disciplines; this reflects its broad appeal and use­fulness as one of the steps in exploratory data analysis. Due to the increasing amount of data available online, the World Wide Web has become one of the most valuable resources for information retrievals and knowledge discoveries. Web mining technologies are the right solutions for knowledge discovery on the Web. In this dissertation, we focus on web page clustering based on their content. A web page clustering system can be useful in web search for grouping search results into closely related sets of documents. It can improve similarity search by focusing on sets of relevant documents. At the same time, web page clustering is much more difficult than pure-text clustering due to a large variety of noisy information embedded in web pages. In this dissertation, web page clustering problem is addressed by the technique in­spired by cemetery organization behavior of ants. Impact of dimensionality reduction and feature selection is also studied.