Identifying Multiple Authors in a Binary Program

Meng, Xiaozhu; Miller, Barton P.; Jun, Kwang-Sung

doi:10.1007/978-3-319-66399-9_16

Cited by 38 publications

(38 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two approaches are considered to identify a programmer of software based on the available code form, which are source code and executable binaries. Several techniques for identifying programmers when only executable machine code is available [18,39,48]. On the other hand, most of the works on code authorship identification are conducted on source code [14,16,17,24,25,29,[35][36][37]45].…”

Section: Single-author Code Identificationmentioning

confidence: 99%

“…However, most of those approaches are inapplicable to authorship identification on structured code due to the inflexibility of the written code expressions established by the syntax rules. Recently, several efforts have investigated authorship identification of structured code, such as computer programs [1,16,17,39]. These efforts established that programmers have distinctive and identifying programming styles that can survive compilation [18,39].…”

Section: Introductionmentioning

confidence: 99%

“…Code authorship identification has many useful applications and can be either binary-or source codebased identification. Binary-based techniques [5,18,39,48] are applicable to applications such as malware, proprietary software, and legacy code [39]. Source codebased techniques are applicable when the source code is available, e.g., in software copyright infringement [28], code authorship disputes [59], plagiarism detection [15], and code integrity investigations [38].…”

Section: Introductionmentioning

confidence: 99%

“…Most of the existing code authorship identification techniques assume a single author per code sample, an assumption that does not always hold. For example, modern software projects are often the result of collaborative efforts, even with malware development due to shared code [39]. As a result, a multi-author and fine-grained identification from small segments of code within a single file becomes essential, especially for forensic applications.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multi-χ: Identifying Multiple Authors from Source Code Files

Abuhamad

Abuhmed

Nyang

et al. 2020

Proceedings on Privacy Enhancing Technologies

View full text Add to dashboard Cite

Most authorship identification schemes assume that code samples are written by a single author. However, real software projects are typically the result of a team effort, making it essential to consider a finegrained multi-author identification in a single code sample, which we address with Multi-χ. Multi-χ leverages a deep learning-based approach for multi-author identification in source code, is lightweight, uses a compact representation for efficiency, and does not require any code parsing, syntax tree extraction, nor feature selection. In Multi-χ, code samples are divided into small segments, which are then represented as a sequence of n-dimensional term representations. The sequence is fed into an RNN-based verification model to assist a segment integration process which integrates positively verified segments, i.e., integrates segments that have a high probability of being written by one author. Finally, the resulting segments from the integration process are represented using word2vec or TF-IDF and fed into the identification model. We evaluate Multi-χ with several Github projects (Caffe, Facebook’s Folly, Tensor-Flow, etc.) and show remarkable accuracy. For example, Multi-χ achieves an authorship example-based accuracy (A-EBA) of 86.41% and per-segment authorship identification of 93.18% for identifying 562 programmers. We examine the performance against multiple dimensions and design choices, and demonstrate its effectiveness.

show abstract

Section: Single-author Code Identificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-χ: Identifying Multiple Authors from Source Code Files

Abuhamad

Abuhmed

Nyang

et al. 2020

Proceedings on Privacy Enhancing Technologies

View full text Add to dashboard Cite

show abstract

“…According to existing research [28], [21], [4], they can identify hundreds of authors with high accuracy from program binaries. Although those results are scientifically impressive, this approach faces a big problem if they are applied in some real forensics scenario such as malware analysis.…”

Section: Introductionmentioning

confidence: 99%