Analysis of Efficient Classification Algorithms in Web Mining

Abstract

Education is a tremendous and critical concern, in current years the amount of data stored in academic database is developing rapidly. The protected database consists of hidden data about the development of the scholar’s performance and behavior. The ability to assess the faculties overall performance annually could be very critical in academic environments. Web content mining is the method of extracting useful data from the contents of web pages. Content records are the collection of data that a web page is designed to hold. Content data consist of text, photograph, audio, video, hyperlink or dependent data which includes lists or tables. From this method, the scholars’ behavior is analyzed by using the Internet log files, which holds the data such as staffs searched web pages and their history was extracted for analysis. The analysis of staffs’ behavior in the learning and teaching environment is based on log files created on the server during the course of interaction between learners and the electronic syllabus. This study emphasizes on concept of various classification algorithms, for processing the data available in the web. There are numerous algorithms for classification, among those three major algorithms such as Naïve Bayes algorithm, Support Vector Machine algorithm (SVM) and finally Artificial Neural Network algorithms (ANN) were considered for analysis. These three algorithms were used for analyzing the web log files, where the data to be handles resides. The results of all the three algorithms have been clearly defined, in order to choose the efficient classification algorithm.