Multi-Label Classification of Pure Code
Currently, there is a significant amount of public code in the IT communities, programming forums and code repositories. Many of these codes lack classification labels, or have imprecise labels, which causes inconvenience to code management and retrieval. Some classification methods have been propos...
Published in: | INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING |
---|---|
Main Authors: | , , , |
Format: | Article; Early Access |
Language: | English |
Published: |
WORLD SCIENTIFIC PUBL CO PTE LTD
2024
|
Subjects: | |
Online Access: | https://www-webofscience-com.uitm.idm.oclc.org/wos/woscc/full-record/WOS:001278652300001 |
author |
Gao Bin; Qin Hongwu; Ma Xiuqin |
---|---|
spellingShingle |
Gao Bin; Qin Hongwu; Ma Xiuqin Multi-Label Classification of Pure Code Computer Science; Engineering |
author_facet |
Gao Bin; Qin Hongwu; Ma Xiuqin |
author_sort |
Gao |
spelling |
Gao, Bin; Qin, Hongwu; Ma, Xiuqin Multi-Label Classification of Pure Code INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING English Article; Early Access Currently, there is a significant amount of public code in the IT communities, programming forums and code repositories. Many of these codes lack classification labels, or have imprecise labels, which causes inconvenience to code management and retrieval. Some classification methods have been proposed to automatically assign labels to the code. However, these methods mainly rely on code comments or surrounding text, and the classification effect is limited by the quality of them. So far, there are a few methods that rely solely on the code itself to assign labels to the code. In this paper, an encoder-only method is proposed to assign multiple labels to the code of an algorithmic problem, in which UniXcoder is employed to encode the input code and the encoding results correspond to the output labels through the classification heads. The proposed method relies only on the code itself. We construct a dataset to evaluate the proposed method, which consists of source code in three programming languages (C++, Java, Python) with a total size of approximately 120K. The results of the comparative experiment show that the proposed method has better performance in multi-label classification task of pure code than encoder-decoder methods. WORLD SCIENTIFIC PUBL CO PTE LTD 0218-1940 1793-6403 2024 10.1142/S0218194024500311 Computer Science; Engineering WOS:001278652300001 https://www-webofscience-com.uitm.idm.oclc.org/wos/woscc/full-record/WOS:001278652300001 |
title |
Multi-Label Classification of Pure Code |
title_short |
Multi-Label Classification of Pure Code |
title_full |
Multi-Label Classification of Pure Code |
title_fullStr |
Multi-Label Classification of Pure Code |
title_full_unstemmed |
Multi-Label Classification of Pure Code |
title_sort |
Multi-Label Classification of Pure Code |
container_title |
INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING |
language |
English |
format |
Article; Early Access |
description |
Currently, there is a significant amount of public code in the IT communities, programming forums and code repositories. Many of these codes lack classification labels, or have imprecise labels, which causes inconvenience to code management and retrieval. Some classification methods have been proposed to automatically assign labels to the code. However, these methods mainly rely on code comments or surrounding text, and the classification effect is limited by the quality of them. So far, there are a few methods that rely solely on the code itself to assign labels to the code. In this paper, an encoder-only method is proposed to assign multiple labels to the code of an algorithmic problem, in which UniXcoder is employed to encode the input code and the encoding results correspond to the output labels through the classification heads. The proposed method relies only on the code itself. We construct a dataset to evaluate the proposed method, which consists of source code in three programming languages (C++, Java, Python) with a total size of approximately 120K. The results of the comparative experiment show that the proposed method has better performance in multi-label classification task of pure code than encoder-decoder methods. |
publisher |
WORLD SCIENTIFIC PUBL CO PTE LTD |
issn |
0218-1940 1793-6403 |
publishDate |
2024 |
container_volume |
|
container_issue |
|
doi_str_mv |
10.1142/S0218194024500311 |
topic |
Computer Science; Engineering |
topic_facet |
Computer Science; Engineering |
accesstype |
|
id |
WOS:001278652300001 |
url |
https://www-webofscience-com.uitm.idm.oclc.org/wos/woscc/full-record/WOS:001278652300001 |
record_format |
wos |
collection |
Web of Science (WoS) |
_version_ |
1809679296836403200 |