Automatic library categorization

Camilo Ernesto Velazquez Rodriguez; Coen De Roover

doi:10.1145/3387940.3392186

Abstract

Software ecosystems contain several types of artefacts such as libraries, documentation and source code files. Recent studies show that the Maven software ecosystem alone already contains over 2.8 million artefacts and over 70, 000 libraries. Given the size of the ecosystem, selecting a library represents a challenge to its users.

The MVNRepository website offers a category-based search func- tionality as a solution. However, not all of the libraries have been categorised, which leads to incomplete search results. This work proposes an approach to the automatic categorisation of libraries through machine learning classifiers trained on class and method names. Our preliminary results show that the approach is accurate, suggesting that large-scale applications may be feasible.

Original language	English
Pages	733-734
Number of pages	2
DOIs	https://doi.org/10.1145/3387940.3392186
Publication status	Published - 2020
Event	3rd International Workshop on Software Health - , Korea, Republic of Duration: 5 Oct 2020 → 11 Oct 2020

Workshop

Workshop	3rd International Workshop on Software Health
Country	Korea, Republic of
Period	5/10/20 → 11/10/20

Keywords

Software Ecosystems
API Category
Text Classification

Access to Document

10.1145/3387940.3392186

FWOEOS10: Automated Assistance for Developing Software in Ecosystems of the Future
De Roover, C., Mens, T., Demeyer, S. & Cleve, A.
1/01/18 → 31/12/21
Project: Fundamental

Cite this

@conference{28f171f75218419eb759b590aac0123c,

title = "Automatic library categorization",

abstract = "Software ecosystems contain several types of artefacts such as libraries, documentation and source code files. Recent studies show that the Maven software ecosystem alone already contains over 2.8 million artefacts and over 70, 000 libraries. Given the size of the ecosystem, selecting a library represents a challenge to its users.The MVNRepository website offers a category-based search func- tionality as a solution. However, not all of the libraries have been categorised, which leads to incomplete search results. This work proposes an approach to the automatic categorisation of libraries through machine learning classifiers trained on class and method names. Our preliminary results show that the approach is accurate, suggesting that large-scale applications may be feasible.",

keywords = "Software Ecosystems, API Category, Text Classification",

author = "{Velazquez Rodriguez}, {Camilo Ernesto} and {De Roover}, Coen",

year = "2020",

doi = "10.1145/3387940.3392186",

language = "English",

pages = "733--734",

note = "3rd International Workshop on Software Health ; Conference date: 05-10-2020 Through 11-10-2020",

}

TY - CONF

T1 - Automatic library categorization

AU - Velazquez Rodriguez, Camilo Ernesto

AU - De Roover, Coen

PY - 2020

Y1 - 2020

N2 - Software ecosystems contain several types of artefacts such as libraries, documentation and source code files. Recent studies show that the Maven software ecosystem alone already contains over 2.8 million artefacts and over 70, 000 libraries. Given the size of the ecosystem, selecting a library represents a challenge to its users.The MVNRepository website offers a category-based search func- tionality as a solution. However, not all of the libraries have been categorised, which leads to incomplete search results. This work proposes an approach to the automatic categorisation of libraries through machine learning classifiers trained on class and method names. Our preliminary results show that the approach is accurate, suggesting that large-scale applications may be feasible.

AB - Software ecosystems contain several types of artefacts such as libraries, documentation and source code files. Recent studies show that the Maven software ecosystem alone already contains over 2.8 million artefacts and over 70, 000 libraries. Given the size of the ecosystem, selecting a library represents a challenge to its users.The MVNRepository website offers a category-based search func- tionality as a solution. However, not all of the libraries have been categorised, which leads to incomplete search results. This work proposes an approach to the automatic categorisation of libraries through machine learning classifiers trained on class and method names. Our preliminary results show that the approach is accurate, suggesting that large-scale applications may be feasible.

KW - Software Ecosystems

KW - API Category

KW - Text Classification

UR - http://www.scopus.com/inward/record.url?scp=85093121571&partnerID=8YFLogxK

UR - https://soft.vub.ac.be/Publications/2020/vub-tr-soft-20-06.pdf

U2 - 10.1145/3387940.3392186

DO - 10.1145/3387940.3392186

M3 - Unpublished abstract

SP - 733

EP - 734

T2 - 3rd International Workshop on Software Health

Y2 - 5 October 2020 through 11 October 2020

ER -

Automatic library categorization

Abstract

Workshop

Keywords

Access to Document

Fingerprint

Projects

FWOEOS10: Automated Assistance for Developing Software in Ecosystems of the Future

Cite this