Comparative Study of Back - Propagation and Monte - Carlo Artificial Neural Network for Plagiarism Detection in Nepali Language

Abstract

This research work develops two frameworks for detecting plagiarism of Nepali language literatures incorporating Monte Carlo based Artificial Neural Network (MCANN) and Backpropagation (BP) neural network, which was applied for the plagiarism detection on certain document type segment. Neural Network training is considered using Monte Carlo based family of algorithms as of these algorithms superiority and robustness. Both the frameworks are tested on two different datasets and results were analysed and discussed. Convergence of MCANN is faster in comparison to traditional BP algorithm. MCANN algorithm achieve a convergence in the range of 10 2 to 10 7 for the training error in 40 epochs while general BP algorithm is unable to achieve such a convergence even in 400 epochs. Also, the mean accuracy of BP and MCANN are respectively found to be in the range of 98.657 and 99.864 during paragraph based and line based comparison of the documents. Thus, MCANN is efficient for plagiarism detection in comparison to BP for Nepali language documents.