Code Clone Detection Using COCOMO-1

Ekta Manhas, Samriti Rana

Abstract: Now adays, Copy and Paste of code fragments has been regularly practiced in development of software. Because of limitations of time and lack of knowledge, the programmers use the code strategy known as code cloning. Clones may cause many problems that are the probability of errors and the maintenance cost get increased. The modifications become difficult because of clones. Therefore, the detection and removal of clones is necessary. It has been observed that lot of tools, techniques and classifiers have been already tried in the concept of textual parameter code cloning detection, but there are chances of improvement of the accuracy pattern of classification in code cloning. So, the motive of this research work is to enhance the accuracy of the system with reduction of error rate using neural network classifier. In the proposed work, ANN would be used that involves training of data. Utilization of ANN has been done for code cloning as it provides good training tool with robust results. The analysis has been carried out in MATLAB environment and the metrics like accuracy and error rate are calculated. The accuracy upto 95 % have been obtained. Keywords: Code clone, COCOMO-II, Neural Network, Accuracy, Error rate

Title:	Code Clone Detection Using COCOMO-1
Author Name:	Ekta Manhas, Samriti Rana
Abstract:	Abstract: Now adays, Copy and Paste of code fragments has been regularly practiced in development of software. Because of limitations of time and lack of knowledge, the programmers use the code strategy known as code cloning. Clones may cause many problems that are the probability of errors and the maintenance cost get increased. The modifications become difficult because of clones. Therefore, the detection and removal of clones is necessary. It has been observed that lot of tools, techniques and classifiers have been already tried in the concept of textual parameter code cloning detection, but there are chances of improvement of the accuracy pattern of classification in code cloning. So, the motive of this research work is to enhance the accuracy of the system with reduction of error rate using neural network classifier. In the proposed work, ANN would be used that involves training of data. Utilization of ANN has been done for code cloning as it provides good training tool with robust results. The analysis has been carried out in MATLAB environment and the metrics like accuracy and error rate are calculated. The accuracy upto 95 % have been obtained. Keywords: Code clone, COCOMO-II, Neural Network, Accuracy, Error rate
Cite this article:	Ekta Manhas, Samriti Rana , " Code Clone Detection Using COCOMO-1 " , International Journal of Emerging Trends & Technology in Computer Science (IJETTCS), Volume 6, Issue 5, September - October 2017 , pp. 190-195 , ISSN 2278-6856.
Full Text [PDF] Home