Flowchart-Based Cross-Language Source Code Similarity Detection

Abstract

Source code similarity detection has various applications in code plagiarism detection and software intellectual property protection. In computer programming teaching, students may convert the source code written in one programming language into another language for their code assignment submission. Existing similarity measures of source code written in the same language are not applicable for the cross-language code similarity detection because of syntactic differences among different programming languages. Meanwhile, existing cross-language source similarity detection approaches are susceptible to complex code obfuscation techniques, such as replacing equivalent control structure and adding redundant statements. To solve this problem, we propose a cross-language code similarity detection (CLCSD) approach based on code flowcharts. In general, two source code fragments written in different programming languages are transformed into standardized code flowcharts (SCFC), and their similarity is obtained by measuring their corresponding SCFC. More specifically, we first introduce the standardized code flowchart (SCFC) model to be the uniform flowcharts representation of source code written in different languages. SCFC is language-independent, and therefore, it can be used as the intermediate structure for source code similarity detection. Meanwhile, transformation techniques are given to transform source code written in a specific programming language into an SCFC. Second, we propose the SCFC-SPGK algorithm based on the shortest path graph kernel to measure the similarity between two SCFCs. Thus, the similarity between two pieces of source code in different programming languages is given by the similarity between SCFCs. Experimental results show that compared with existing approaches, CLCSD has higher accuracy in cross-language source code similarity detection. Furthermore, CLCSD cannot only handle common source code obfuscation techniques used by students in computer programming teaching but also obtain nearly 90% accuracy in dealing with some complex obfuscation techniques.

[Retracted] Remote Judgment Method of Painting Image Style Plagiarism Based on Wireless Network Multitask Learning

“…It can be seen from the above figure that when the false alarm rate is the same, the discrimination rate of this method is always better than the other five methods proposed in References [5][6][7][8][9]. When the false alarm rate reaches 1%, the discrimination rate of the method in this paper is close to 82%, the discrimination rate of the method in Reference [7] is 43%, the discrimination rate of the method in Reference [8] is 57%, and the discrimination rate of the method in Reference [9] is 70% (63% and 61% for References [5,6], respectively). Therefore, the method proposed in this paper has the best discrimination effect performance.…”

“…In the formula, IR represents the discrimination rate, DR represents the number of plagiarism incidents of the identified painting image, and S represents the number of actual painting image incidents. In order to verify the performance of the method proposed in this paper, we do the experiment as well as with the other five methods proposed in References [5][6][7][8][9], and the results of IR-FIR are shown in Figure 5.…”

“…A comprehensive analysis of the experimental data in Figure 6 shows the forecast index of the six prediction methods. In order to verify the performance, we have given the performance of FI as well as compare the method proposed in this paper with methods proposed in References [5][6][7][8][9]. Among the six methods for identifying plagiarism in painting images, the indicator of the proposed method is the lowest, which fully shows that the proposed method can achieve very satisfactory results.…”

“…Reference [5] discusses the painting art style conversion technology in NPR, uses image representation derived from convolutional neural networks, and designs an art style algorithm. The algorithm can separate and recombine the content and style of the image and can produce new images with high quality.…”

“…Method proposed in this paper Method in reference [9] Method in reference [5] Method in reference [6] Method in reference [8] Method in reference [7] where FI is forecast index, M is the sum of MSPE of N times, and N is the number of experiment. Figure 6 gives the comparison of the forecast index value of different methods.…”