Building Implicit Vector Representations of Individual Coding Style

Standard

Building Implicit Vector Representations of Individual Coding Style. / Kovalenko, Vladimir; Bogomolov, Egor; Bryksin, Timofey; Bacchelli, Alberto.

2020. 117-124.

Research output: Contribution to conference › Paper › peer-review

BibTeX

@conference{530d1cdfa43f4e79aad726f116a42212,

title = "Building Implicit Vector Representations of Individual Coding Style",

abstract = "We propose a new approach to building vector representations of individual developers by capturing their individual contribution style, or coding style. Such representations can find use in the next generation of software development team collaboration tools, for example by enabling the tools to track knowledge transfer in teams. The key idea of our approach is to avoid using explicitly defined metrics of coding style and instead build the representations through training a model for authorship recognition and extracting the representations of individual developers from the trained model. By empirically evaluating the output of our approach, we find that implicitly built individual representations reflect some properties of team structure: developers who report learning from each other are represented closer to each other.",

author = "Vladimir Kovalenko and Egor Bogomolov and Timofey Bryksin and Alberto Bacchelli",

year = "2020",

month = jun,

day = "27",

doi = "10.1145/3387940.3391494",

language = "English",

pages = "117--124",

}

RIS

TY - CONF

T1 - Building Implicit Vector Representations of Individual Coding Style

AU - Kovalenko, Vladimir

AU - Bogomolov, Egor

AU - Bryksin, Timofey

AU - Bacchelli, Alberto

PY - 2020/6/27

Y1 - 2020/6/27

N2 - We propose a new approach to building vector representations of individual developers by capturing their individual contribution style, or coding style. Such representations can find use in the next generation of software development team collaboration tools, for example by enabling the tools to track knowledge transfer in teams. The key idea of our approach is to avoid using explicitly defined metrics of coding style and instead build the representations through training a model for authorship recognition and extracting the representations of individual developers from the trained model. By empirically evaluating the output of our approach, we find that implicitly built individual representations reflect some properties of team structure: developers who report learning from each other are represented closer to each other.

AB - We propose a new approach to building vector representations of individual developers by capturing their individual contribution style, or coding style. Such representations can find use in the next generation of software development team collaboration tools, for example by enabling the tools to track knowledge transfer in teams. The key idea of our approach is to avoid using explicitly defined metrics of coding style and instead build the representations through training a model for authorship recognition and extracting the representations of individual developers from the trained model. By empirically evaluating the output of our approach, we find that implicitly built individual representations reflect some properties of team structure: developers who report learning from each other are represented closer to each other.

UR - https://www.mendeley.com/catalogue/a7f467ba-c598-3045-806f-fc92e78f7151/

UR - http://www.scopus.com/inward/record.url?scp=85093109954&partnerID=8YFLogxK

U2 - 10.1145/3387940.3391494

DO - 10.1145/3387940.3391494

M3 - Paper

SP - 117

EP - 124

ER -

ID: 64761981