Towards Improving Code Stylometry Analysis in Underground Forums

Code Stylometry has emerged as a powerful mechanism to identify programmers. While there have been significant advances in the field, existing mechanisms underperform in challenging domains. One such domain is studying the provenance of code shared in underground forums, where code posts tend to hav...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Tereszkowski-Kaminski Michal, Pastrana Sergio, Blasco Jorge, Suarez-Tangil Guillermo
Formato:	article
Lenguaje:	EN
Publicado:	Sciendo 2022
Materias:	authorship attribution underground forums language selection code clone detection Ethics BJ1-1725 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/0f09d9231f5b412abab528bb2857f5c3
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:0f09d9231f5b412abab528bb2857f5c3
record_format	dspace
spelling	oai:doaj.org-article:0f09d9231f5b412abab528bb2857f5c32021-12-05T14:11:10ZTowards Improving Code Stylometry Analysis in Underground Forums2299-098410.2478/popets-2022-0007https://doaj.org/article/0f09d9231f5b412abab528bb2857f5c32022-01-01T00:00:00Zhttps://doi.org/10.2478/popets-2022-0007https://doaj.org/toc/2299-0984Code Stylometry has emerged as a powerful mechanism to identify programmers. While there have been significant advances in the field, existing mechanisms underperform in challenging domains. One such domain is studying the provenance of code shared in underground forums, where code posts tend to have small or incomplete source code fragments. This paper proposes a method designed to deal with the idiosyncrasies of code snippets shared in these forums. Our system fuses a forum-specific learning pipeline with Conformal Prediction to generate predictions with precise confidence levels as a novelty. We see that identifying unreliable code snippets is paramount to generate high-accuracy predictions, and this is a task where traditional learning settings fail. Overall, our method performs as twice as well as the state-of-the-art in a constrained setting with a large number of authors (i.e., 100). When dealing with a smaller number of authors (i.e., 20), it performs at high accuracy (89%). We also evaluate our work on an open-world assumption and see that our method is more effective at retaining samples.Tereszkowski-Kaminski MichalPastrana SergioBlasco JorgeSuarez-Tangil GuillermoSciendoarticleauthorship attributionunderground forumslanguage selectioncode clone detectionEthicsBJ1-1725Electronic computers. Computer scienceQA75.5-76.95ENProceedings on Privacy Enhancing Technologies, Vol 2022, Iss 1, Pp 126-147 (2022)
institution	DOAJ
collection	DOAJ
language	EN
topic	authorship attribution underground forums language selection code clone detection Ethics BJ1-1725 Electronic computers. Computer science QA75.5-76.95
spellingShingle	authorship attribution underground forums language selection code clone detection Ethics BJ1-1725 Electronic computers. Computer science QA75.5-76.95 Tereszkowski-Kaminski Michal Pastrana Sergio Blasco Jorge Suarez-Tangil Guillermo Towards Improving Code Stylometry Analysis in Underground Forums
description	Code Stylometry has emerged as a powerful mechanism to identify programmers. While there have been significant advances in the field, existing mechanisms underperform in challenging domains. One such domain is studying the provenance of code shared in underground forums, where code posts tend to have small or incomplete source code fragments. This paper proposes a method designed to deal with the idiosyncrasies of code snippets shared in these forums. Our system fuses a forum-specific learning pipeline with Conformal Prediction to generate predictions with precise confidence levels as a novelty. We see that identifying unreliable code snippets is paramount to generate high-accuracy predictions, and this is a task where traditional learning settings fail. Overall, our method performs as twice as well as the state-of-the-art in a constrained setting with a large number of authors (i.e., 100). When dealing with a smaller number of authors (i.e., 20), it performs at high accuracy (89%). We also evaluate our work on an open-world assumption and see that our method is more effective at retaining samples.
format	article
author	Tereszkowski-Kaminski Michal Pastrana Sergio Blasco Jorge Suarez-Tangil Guillermo
author_facet	Tereszkowski-Kaminski Michal Pastrana Sergio Blasco Jorge Suarez-Tangil Guillermo
author_sort	Tereszkowski-Kaminski Michal
title	Towards Improving Code Stylometry Analysis in Underground Forums
title_short	Towards Improving Code Stylometry Analysis in Underground Forums
title_full	Towards Improving Code Stylometry Analysis in Underground Forums
title_fullStr	Towards Improving Code Stylometry Analysis in Underground Forums
title_full_unstemmed	Towards Improving Code Stylometry Analysis in Underground Forums
title_sort	towards improving code stylometry analysis in underground forums
publisher	Sciendo
publishDate	2022
url	https://doaj.org/article/0f09d9231f5b412abab528bb2857f5c3
work_keys_str_mv	AT tereszkowskikaminskimichal towardsimprovingcodestylometryanalysisinundergroundforums AT pastranasergio towardsimprovingcodestylometryanalysisinundergroundforums AT blascojorge towardsimprovingcodestylometryanalysisinundergroundforums AT suareztangilguillermo towardsimprovingcodestylometryanalysisinundergroundforums
_version_	1718371280902684672

Towards Improving Code Stylometry Analysis in Underground Forums

Ejemplares similares