Software defect prediction via LSTM

Software quality plays an important role in the software lifecycle. Traditional software defect prediction approaches mainly focused on using hand-crafted features to detect defects. However, like human languages, programming languages contain rich semantic and structural information, and the cause...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IET software Ročník 14; číslo 4; s. 443 - 450
Hlavní autoři: Deng, Jiehan, Lu, Lu, Qiu, Shaojian
Médium: Journal Article
Jazyk:angličtina
Vydáno: The Institution of Engineering and Technology 01.08.2020
Témata:
AST node sequence
contextual features
defective code
feature extraction
human languages
learning (artificial intelligence)
long short-term memory network
LSTM
machine learning techniques
numerical vectors
open source projects
program abstract syntax trees
program debugging
program diagnostics
programming languages
public domain software
recurrent neural nets
Research Article
semantic features
software defect prediction approaches
software lifecycle
software quality
structural information
trees (mathematics)
word embedding techniques
program debugging
open source projects
long short-term memory network
contextual features
LSTM
defective code
public domain software
program diagnostics
AST node sequence
human languages
recurrent neural nets
trees (mathematics)
software lifecycle
software defect prediction approaches
software quality
machine learning techniques
programming languages
word embedding techniques
program abstract syntax trees
feature extraction
structural information
learning (artificial intelligence)
semantic features
numerical vectors
ISSN:1751-8806, 1751-8814, 1751-8814
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Software quality plays an important role in the software lifecycle. Traditional software defect prediction approaches mainly focused on using hand-crafted features to detect defects. However, like human languages, programming languages contain rich semantic and structural information, and the cause of defective code is closely related to its context. Failing to catch this significant information, the performance of traditional approaches is far from satisfactory. In this study, the authors leveraged a long short-term memory (LSTM) network to automatically learn the semantic and contextual features from the source code. Specifically, they first extract the program's Abstract Syntax Trees (ASTs), which is made up of AST nodes, and then evaluate what and how much information they can preserve for several node types. They traverse the AST of each file and fed them into the LSTM network to automatically the semantic and contextual features of the program, which is then used to determine whether the file is defective. Experimental results on several opensource projects showed that the proposed LSTM method is superior to the state-of-the-art methods.
ISSN:1751-8806
1751-8814
1751-8814
DOI:10.1049/iet-sen.2019.0149