Source Code Authorship Attribution Using Hybrid Approach of Program Dependence Graph and Deep Learning Model

Ullah, Farhan; Wang, Junfeng; Jabbar, Sohail; Al‐Turjman, Fadi; Alazab, Mamoun

doi:10.1109/access.2019.2943639

Cited by 33 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also various modifications of these algorithms that use treeweighting principles [14] to improve detection or deal with efficient comparison of syntax trees using fingerprinting [15] or hashing [16]. Systems that use machine learning, such as neural networks, are also innovative [17], [18].…”

Section: Current Statementioning

confidence: 99%

Abstract Syntax Tree Based Source Code Antiplagiarism System for Large Projects Set

et al. 2020

View full text Add to dashboard Cite

The paper deals with the issue of detecting plagiarism in source code, which we unfortunately encounter when teaching subjects dealing with programming and software development. Many students want to simplify the completion of the course and therefore submit modified source codes of their classmates or even those found on the Internet. Some try to modify the source code e.g. by changing the identifiers of classes, methods and variables to different ones, by changing the corresponding loops, by introducing new methods or by changing the order of methods in the source code or in other ways. We focused directly on this problem and designed our own anti-plagiarism system that we describe in this paper. The designed system consists of three parts during which the source code is processed using six designed algorithms. The basis is the processing of the source code and its transformation into an abstract syntax tree, consisting of two types of nodes, which is then vectorized using our modified DECKARD algorithm. The vectors are then clustered and stored in a database from which similar parts of the source code can be searched. The output of the system is then the final report containing a list of matches with similarities of all works that have been added to the database until then. The designed anti-plagiarism system is finally compared with the success of plagiarism detection performed by the two most used anti-plagiarism tools, namely JPlag and MOSS. It is evaluated on assignments elaborated by students from the courses dealing with object-oriented programming at our faculty.

show abstract

Section: Current Statementioning

confidence: 99%

Abstract Syntax Tree Based Source Code Antiplagiarism System for Large Projects Set

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Ullah et al [6] used a program dependence graph (PDG) along with the deep learning model to identify the authors from the source code of different languages. First, the PDG is used to extract the control flow and data variation features from source code files.…”

Section: Related Workmentioning

confidence: 99%

“…Numerous studies [4]- [6] have been conducted to address the source code authorship attribution problem. The approaches usually focus on sophisticated graph-based features such as abstract syntax trees, program dependency graphs, and machine learning algorithms.…”

Section: Introductionmentioning

confidence: 99%

Language and Obfuscation Oblivious Source Code Authorship Attribution

et al. 2020

View full text Add to dashboard Cite

present an efficient approach to learn a novel representation using deep metric learning. The existing state of the art approaches tokenize the source code and work on the keyword level, limiting the elements of style they can consider. Our approach uses the raw character stream of source code. It can examine keywords and different stylistic features such as variable naming conventions or using tabs vs. spaces, enabling us to learn a richer representation than other keyword-based approaches. Our approach uses a character-level Convolutional Neural Network (CNN). We train the CNN to map the input character stream to a dense vector, mapping the source codes authored by the same author close to each other. In contrast, source codes written by different programmers are mapped farther apart in the embedding space. We then feed these source code vectors into the K-nearest neighbor (KNN) classifier that uses Manhattan-distance to perform authorship attribution. We validated our approach on Google Code Jam (GCJ) dataset across three different programming languages. We prepare our large-scale dataset in such a way that it does not induce type-I error. Our approach is more scalable and efficient than existing methods. We were able to achieve an accuracy of 84.94% across 20,458 authors, which is more than twice the scale of any previous study under a much more challenging setting.

show abstract

“…On the opposite side, code initiation attribution assumes an important job in software forensics activities, security examination, and software plagiarism detection [15] particularly for focusing on malware creators. The malware writers compose malicious software that can co-operate the compilation method in the PC system [16,17].…”

Section: Introductionmentioning

confidence: 99%

Cybersecurity Threats Detection In IoT Using Krill Based Deep Neural Network Stacked Auto Encoders

Rana¹,

Chauhan

Patil³

2021

Preprint

View full text Add to dashboard Cite

The Internet of things (IoT) has concerned much significance for some manufacturing sectors including clinical fields, co-ordinations following, savvy urban communities, and automobiles. Anyway as a worldview, it is sensitive to different sorts of cyber-attacks. Customary very good quality security resolutions for guarantee an IoT structure are not reasonable. This deduces clever organization-based security plans as AI arrangements ought to be made. In this work, we propose Cyber Security Threats recognition in IoT utilizing Krill Based Deep Neural Network Stacked Auto Encoders (KDNN-SAE). In our proposed approach, first, the information pre-processing measure was acted in the underlying development before isolating the dataset into two segments: preparing and test. At that point, flow-based features are extracted from the pre-processed information. By then, the properties to be utilized by the algorithms are chosen in the attribute determination utilizing the Genetic Algorithm (GA). At last, our methodology completes with the execution of the machine learning algorithm KDNN-SAE. The exploratory results show that the introduced method beats the existing techniques to different execution measures.

show abstract

Source Code Authorship Attribution Using Hybrid Approach of Program Dependence Graph and Deep Learning Model

Cited by 33 publications

References 33 publications

Abstract Syntax Tree Based Source Code Antiplagiarism System for Large Projects Set

Abstract Syntax Tree Based Source Code Antiplagiarism System for Large Projects Set

Language and Obfuscation Oblivious Source Code Authorship Attribution

Cybersecurity Threats Detection In IoT Using Krill Based Deep Neural Network Stacked Auto Encoders

Contact Info

Product

Resources

About