 |
    |
Assignment Copy Detection Using Neuro-genetic Hybrids
Seung-Jin Yang, Yong-Geon Kim, Yung-Keun Kwon, and Byung-Ro Moon
School of Computer Science and Engineering
Seoul National University
Shilim-dong, Kwanak-gu
Seoul, 151-742 Korea
{sjyang,dvp,kwon,moon}@soar.snu.ac.kr
Abstract. It is difficult for teachers to find out copies in their program assignments. We propose a method which detects the similarity of assignments. We first transform a program into a sequence of tokens and provide it to an artificial neural network. The neural network is optimized by a hybrid genetic algorithm. Experimental results showed considerable improvement on artificial neural networks.
LNCS 2724, p. 2426 ff.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2003