An Automated Approach to Domain-specific Term Extraction for Programming Language

Zilie Wang; Botong Zhu; Huobin Tan

doi:10.1145/3510513.3510518

An Automated Approach to Domain-specific Term Extraction for Programming Language

Zilie Wang
, Botong Zhu
, Huobin Tan^*

^*Corresponding author for this work

School of Software

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Automatic Term Extraction is a common upstream task of both natural language and programming language processing tasks. However, most existing domain-specific term extraction methods are designed for natural language, and do not satisfy demand for programming language. In this paper we proposed an automated cross-language method for extracting domain-specific terms from source code. To test the domain-specific terms extraction method on real software projects, we choose open source repositories in Github Collections. Then the extraction results of our method and TF-IDF were checked by crowdsourcing, with 12 experienced software developers. Compared with TF-IDF, our method shows better performance on multiple measures. Especially, the AUC of our method is 0.7050, while TF-IDF is 0.4212.

Original language	English
Title of host publication	ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing
Publisher	Association for Computing Machinery
Pages	25-30
Number of pages	6
ISBN (Electronic)	9781450385848
DOIs	https://doi.org/10.1145/3510513.3510518
State	Published - 10 Dec 2021
Event	10th International Conference on Networks, Communication and Computing, ICNCC 2021 - Virtual, Online, China Duration: 10 Dec 2021 → 12 Dec 2021

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	10th International Conference on Networks, Communication and Computing, ICNCC 2021
Country/Territory	China
City	Virtual, Online
Period	10/12/21 → 12/12/21

Keywords

Automatic term extraction
Domain-specific term extraction
Open source software

Access to Document

10.1145/3510513.3510518

Cite this

Wang, Z., Zhu, B., & Tan, H. (2021). An Automated Approach to Domain-specific Term Extraction for Programming Language. In ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing (pp. 25-30). (ACM International Conference Proceeding Series). Association for Computing Machinery . https://doi.org/10.1145/3510513.3510518

@inproceedings{b52edecce435417ea905c718fa99fd6b,

title = "An Automated Approach to Domain-specific Term Extraction for Programming Language",

abstract = "Automatic Term Extraction is a common upstream task of both natural language and programming language processing tasks. However, most existing domain-specific term extraction methods are designed for natural language, and do not satisfy demand for programming language. In this paper we proposed an automated cross-language method for extracting domain-specific terms from source code. To test the domain-specific terms extraction method on real software projects, we choose open source repositories in Github Collections. Then the extraction results of our method and TF-IDF were checked by crowdsourcing, with 12 experienced software developers. Compared with TF-IDF, our method shows better performance on multiple measures. Especially, the AUC of our method is 0.7050, while TF-IDF is 0.4212.",

keywords = "Automatic term extraction, Domain-specific term extraction, Open source software",

author = "Zilie Wang and Botong Zhu and Huobin Tan",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 10th International Conference on Networks, Communication and Computing, ICNCC 2021 ; Conference date: 10-12-2021 Through 12-12-2021",

year = "2021",

month = dec,

day = "10",

doi = "10.1145/3510513.3510518",

language = "英语",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery ",

pages = "25--30",

booktitle = "ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing",

address = "美国",

}

Wang, Z, Zhu, B & Tan, H 2021, An Automated Approach to Domain-specific Term Extraction for Programming Language. in ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing. ACM International Conference Proceeding Series, Association for Computing Machinery , pp. 25-30, 10th International Conference on Networks, Communication and Computing, ICNCC 2021, Virtual, Online, China, 10/12/21. https://doi.org/10.1145/3510513.3510518

An Automated Approach to Domain-specific Term Extraction for Programming Language. / Wang, Zilie; Zhu, Botong; Tan, Huobin.
ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing. Association for Computing Machinery , 2021. p. 25-30 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An Automated Approach to Domain-specific Term Extraction for Programming Language

AU - Wang, Zilie

AU - Zhu, Botong

AU - Tan, Huobin

PY - 2021/12/10

Y1 - 2021/12/10

N2 - Automatic Term Extraction is a common upstream task of both natural language and programming language processing tasks. However, most existing domain-specific term extraction methods are designed for natural language, and do not satisfy demand for programming language. In this paper we proposed an automated cross-language method for extracting domain-specific terms from source code. To test the domain-specific terms extraction method on real software projects, we choose open source repositories in Github Collections. Then the extraction results of our method and TF-IDF were checked by crowdsourcing, with 12 experienced software developers. Compared with TF-IDF, our method shows better performance on multiple measures. Especially, the AUC of our method is 0.7050, while TF-IDF is 0.4212.

AB - Automatic Term Extraction is a common upstream task of both natural language and programming language processing tasks. However, most existing domain-specific term extraction methods are designed for natural language, and do not satisfy demand for programming language. In this paper we proposed an automated cross-language method for extracting domain-specific terms from source code. To test the domain-specific terms extraction method on real software projects, we choose open source repositories in Github Collections. Then the extraction results of our method and TF-IDF were checked by crowdsourcing, with 12 experienced software developers. Compared with TF-IDF, our method shows better performance on multiple measures. Especially, the AUC of our method is 0.7050, while TF-IDF is 0.4212.

KW - Automatic term extraction

KW - Domain-specific term extraction

KW - Open source software

UR - https://www.scopus.com/pages/publications/85130213123

U2 - 10.1145/3510513.3510518

DO - 10.1145/3510513.3510518

M3 - 会议稿件

AN - SCOPUS:85130213123

T3 - ACM International Conference Proceeding Series

SP - 25

EP - 30

BT - ICNCC 2021 - Proceedings of the 2021 10th International Conference on Networks, Communication and Computing

PB - Association for Computing Machinery

T2 - 10th International Conference on Networks, Communication and Computing, ICNCC 2021

Y2 - 10 December 2021 through 12 December 2021

ER -

An Automated Approach to Domain-specific Term Extraction for Programming Language

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this