An effective classification technique for XML documents using hyper parameter optimized classifiers

Abstract

In real world XML data plays a significant role in the application of World Wide Web. Now a days, in research   the data classification in XML document for heterogeneous structure proves to be a challenging task. A number of algorithms are available in XML data classification process. In the existing technique the performance is degraded in the classification process of XML document. In this paper the machine learning technique TSRSA (Tuning Swarm Rapid Swarm Algorithm) is proposed to classify the XML data. First, the elements are extracted by using kernel vector space model. Second, we classify the XML data using the algorithm of TSRSA optimization technique. TSRSO is using hyper parameters to obtain the better classifier. The experiments are demonstrated in the existing technique ELM (Extreme Machine Learning), Standard algorithms (SVM Support Vector Machine, DT-Decision Tree, NB-Navie Bayes, and KNN-K Nearest Neighbor), KPCA-Kernel Principal Component Analysis and KELM Kernel Extreme Machine. In this research the proposed TSRSA algorithms are compared with the existing technique. The various performance parameters are compared with reference to the existing and the proposed model.