An Algorithm for Multi-Domain Website Classification

Abstract

The web is the largest world-wide communication system of computers. The web has local, academic, commercial and government sites. As the types of websites increases in numbers, the cost and accuracy of manual classification became cumbersome and cannot satisfy the increasing internet service demands, thereby automated classification became important for better and more accurate search engine results. Therefore, this research has proposed an algorithm for classifying different websites automatically by using randomly collected textual data from the webpages. This research also contributed ten dictionaries covering different domains and used as training data in the classification process. Finally, the classification was carried out using the proposed and Naïve Bayes algorithms and found the proposed algorithm outperformed on the scale of accuracy by 1.25%. This research suggests that the proposed algorithm could be applied to any number of domains if the related dictionaries are available.