Data and Knowledge Extraction Based on Structure Analysis of Homogeneous Websites

Abstract

Abstract-The World Wide Web includes several types of website applications. Mainly these applications are related to business, organizations, companies, and others. There is a lack to get raw data sets to study the behavior of the internal structure of each type of these websites. Where websites structures include treasure of links, and sub-links, in addition to some embedded features associated with the internal structure of each website. The objective of this paper is to analysis a set of homogeneous websites to establish raw data sets. These data sets can be employed for several research purposes. It also can be used to extract some invisible aspects/features within the structure. Several steps are required to accomplish this objective; first, to propose an algorithm for structure analysis, second, to implement the proposed algorithm as a software tool for the purpose of extraction and establishment of raw data sets (real data set), third, to extrapolate a set of rules or relations from these data sets. This data set can be employed for researches purposes in the field of web structure mining, to estimate important factors related to websites development processes, and websites ranking. The results comprise creation of Oriented Data Sets (ODS) for research purposes and also for deducing a set of features represents a type of new discovered knowledge in this ODS.