Deep Learning Anti-patterns from Code Metrics History. Barbez, A., Khomh, F., & Gu�h�neuc, Y. In Kim, M. & Besz�des, �., editors, Proceedings of the 35<sup>nd</sup> International Conference on Software Maintenance and Evolution (ICSME), pages 114–124, September–October, 2019. IEEE CS Press. 11 pages.![Deep Learning Anti-patterns from Code Metrics History [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex Anti-patterns are poor solutions to recurring design problems. Number of empirical studies have highlighted the negative impact of anti-patterns on software maintenance which motivated the development of various detection techniques. Most of these approaches rely on structural metrics of software systems to identify affected components while others exploit historical information by analyzing co-changes occurring between code components. By relying solely on one aspect of software systems (i.e., structural or historical), existing approaches miss some precious information which limits their performances. In this paper, we propose CAME (Convolutional Analysis of code Metrics Evolution), a deep-learning based approach that relies on both structural and historical information to detect anti-patterns. Our approach exploits historical values of structural code metrics mined from version control systems and uses a Convolutional Neural Network classifier to infer the presence of anti-patterns from this information. We experiment our approach for the widely known God Class anti-pattern and evaluate its performances on three software systems. With the results of our study, we show that: (1) using historical values of source code metrics allows to increase the precision; (2) CAME outperforms existing static machine-learning classifiers; and (3) CAME outperforms existing detection tools.
@INPROCEEDINGS{Barbez19-ICSME-DLforAPs,
AUTHOR = {Antoine Barbez and Foutse Khomh and Yann-Ga�l Gu�h�neuc},
BOOKTITLE = {Proceedings of the 35<sup>nd</sup> International Conference on Software Maintenance and Evolution (ICSME)},
TITLE = {Deep Learning Anti-patterns from Code Metrics History},
YEAR = {2019},
OPTADDRESS = {},
OPTCROSSREF = {},
EDITOR = {Miryung Kim and �rp�d Besz�des},
MONTH = {September--October},
NOTE = {11 pages.},
OPTNUMBER = {},
OPTORGANIZATION = {},
PAGES = {114--124},
PUBLISHER = {IEEE CS Press},
OPTSERIES = {},
OPTVOLUME = {},
KEYWORDS = {Topic: <b>Code and design smells</b>,
Venue: <c>ICSM</c>, Venue: <c>ICSME</c>},
URL = {http://www.ptidej.net/publications/documents/ICSM19.doc.pdf},
PDF = {http://www.ptidej.net/publications/documents/ICSM19.ppt.pdf},
ABSTRACT = {Anti-patterns are poor solutions to recurring design
problems. Number of empirical studies have highlighted the negative
impact of anti-patterns on software maintenance which motivated the
development of various detection techniques. Most of these approaches
rely on structural metrics of software systems to identify affected
components while others exploit historical information by analyzing
co-changes occurring between code components. By relying solely on
one aspect of software systems (i.e., structural or historical),
existing approaches miss some precious information which limits their
performances. In this paper, we propose CAME (Convolutional Analysis
of code Metrics Evolution), a deep-learning based approach that
relies on both structural and historical information to detect
anti-patterns. Our approach exploits historical values of structural
code metrics mined from version control systems and uses a
Convolutional Neural Network classifier to infer the presence of
anti-patterns from this information. We experiment our approach for
the widely known God Class anti-pattern and evaluate its performances
on three software systems. With the results of our study, we show
that: (1) using historical values of source code metrics allows to
increase the precision; (2) CAME outperforms existing static
machine-learning classifiers; and (3) CAME outperforms existing
detection tools.}
}
Downloads: 0
{"_id":"vxK8txkxosyfFTsyG","bibbaseid":"barbez-khomh-guhneuc-deeplearningantipatternsfromcodemetricshistory-2019","author_short":["Barbez, A.","Khomh, F.","Gu�h�neuc, Y."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Antoine"],"propositions":[],"lastnames":["Barbez"],"suffixes":[]},{"firstnames":["Foutse"],"propositions":[],"lastnames":["Khomh"],"suffixes":[]},{"firstnames":["Yann-Ga�l"],"propositions":[],"lastnames":["Gu�h�neuc"],"suffixes":[]}],"booktitle":"Proceedings of the 35<sup>nd</sup> International Conference on Software Maintenance and Evolution (ICSME)","title":"Deep Learning Anti-patterns from Code Metrics History","year":"2019","optaddress":"","optcrossref":"","editor":[{"firstnames":["Miryung"],"propositions":[],"lastnames":["Kim"],"suffixes":[]},{"firstnames":["�rp�d"],"propositions":[],"lastnames":["Besz�des"],"suffixes":[]}],"month":"September–October","note":"11 pages.","optnumber":"","optorganization":"","pages":"114–124","publisher":"IEEE CS Press","optseries":"","optvolume":"","keywords":"Topic: <b>Code and design smells</b>, Venue: <c>ICSM</c>, Venue: <c>ICSME</c>","url":"http://www.ptidej.net/publications/documents/ICSM19.doc.pdf","pdf":"http://www.ptidej.net/publications/documents/ICSM19.ppt.pdf","abstract":"Anti-patterns are poor solutions to recurring design problems. Number of empirical studies have highlighted the negative impact of anti-patterns on software maintenance which motivated the development of various detection techniques. Most of these approaches rely on structural metrics of software systems to identify affected components while others exploit historical information by analyzing co-changes occurring between code components. By relying solely on one aspect of software systems (i.e., structural or historical), existing approaches miss some precious information which limits their performances. In this paper, we propose CAME (Convolutional Analysis of code Metrics Evolution), a deep-learning based approach that relies on both structural and historical information to detect anti-patterns. Our approach exploits historical values of structural code metrics mined from version control systems and uses a Convolutional Neural Network classifier to infer the presence of anti-patterns from this information. We experiment our approach for the widely known God Class anti-pattern and evaluate its performances on three software systems. With the results of our study, we show that: (1) using historical values of source code metrics allows to increase the precision; (2) CAME outperforms existing static machine-learning classifiers; and (3) CAME outperforms existing detection tools.","bibtex":"@INPROCEEDINGS{Barbez19-ICSME-DLforAPs,\r\n AUTHOR = {Antoine Barbez and Foutse Khomh and Yann-Ga�l Gu�h�neuc},\r\n BOOKTITLE = {Proceedings of the 35<sup>nd</sup> International Conference on Software Maintenance and Evolution (ICSME)},\r\n TITLE = {Deep Learning Anti-patterns from Code Metrics History},\r\n YEAR = {2019},\r\n OPTADDRESS = {},\r\n OPTCROSSREF = {},\r\n EDITOR = {Miryung Kim and �rp�d Besz�des},\r\n MONTH = {September--October},\r\n NOTE = {11 pages.},\r\n OPTNUMBER = {},\r\n OPTORGANIZATION = {},\r\n PAGES = {114--124},\r\n PUBLISHER = {IEEE CS Press},\r\n OPTSERIES = {},\r\n OPTVOLUME = {},\r\n KEYWORDS = {Topic: <b>Code and design smells</b>, \r\n Venue: <c>ICSM</c>, Venue: <c>ICSME</c>},\r\n URL = {http://www.ptidej.net/publications/documents/ICSM19.doc.pdf},\r\n PDF = {http://www.ptidej.net/publications/documents/ICSM19.ppt.pdf},\r\n ABSTRACT = {Anti-patterns are poor solutions to recurring design \r\n problems. Number of empirical studies have highlighted the negative \r\n impact of anti-patterns on software maintenance which motivated the \r\n development of various detection techniques. Most of these approaches \r\n rely on structural metrics of software systems to identify affected \r\n components while others exploit historical information by analyzing \r\n co-changes occurring between code components. By relying solely on \r\n one aspect of software systems (i.e., structural or historical), \r\n existing approaches miss some precious information which limits their \r\n performances. In this paper, we propose CAME (Convolutional Analysis \r\n of code Metrics Evolution), a deep-learning based approach that \r\n relies on both structural and historical information to detect \r\n anti-patterns. Our approach exploits historical values of structural \r\n code metrics mined from version control systems and uses a \r\n Convolutional Neural Network classifier to infer the presence of \r\n anti-patterns from this information. We experiment our approach for \r\n the widely known God Class anti-pattern and evaluate its performances \r\n on three software systems. With the results of our study, we show \r\n that: (1) using historical values of source code metrics allows to \r\n increase the precision; (2) CAME outperforms existing static \r\n machine-learning classifiers; and (3) CAME outperforms existing \r\n detection tools.}\r\n}\r\n\r\n","author_short":["Barbez, A.","Khomh, F.","Gu�h�neuc, Y."],"editor_short":["Kim, M.","Besz�des, �."],"key":"Barbez19-ICSME-DLforAPs","id":"Barbez19-ICSME-DLforAPs","bibbaseid":"barbez-khomh-guhneuc-deeplearningantipatternsfromcodemetricshistory-2019","role":"author","urls":{"Paper":"http://www.ptidej.net/publications/documents/ICSM19.doc.pdf"},"keyword":["Topic: <b>Code and design smells</b>","Venue: <c>ICSM</c>","Venue: <c>ICSME</c>"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"http://www.yann-gael.gueheneuc.net/Work/Publications/Biblio/complete-bibliography.bib","dataSources":["8vn5MSGYWB4fAx9Z4"],"keywords":["topic: <b>code and design smells</b>","venue: <c>icsm</c>","venue: <c>icsme</c>"],"search_terms":["deep","learning","anti","patterns","code","metrics","history","barbez","khomh","gu�h�neuc"],"title":"Deep Learning Anti-patterns from Code Metrics History","year":2019}