.png){kind=icon}
返回
Generating class name in sequential manner using convolution attentionneural network
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2022.116854.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Software code comprehension is strongly dependent on identifier names; therefore, software developers spenda lot of time assigning suitable names to identifiers. Manually suggesting a good name is a time takingand hard problem for developers. For automatic identifiers name recommendation, various techniques havebeen proposed. Most of the work has been done for method name prediction. We found very few researchworks on class name recommendation. A good class name communicates the class's intent, whereas bad onescreate confusion and frustration in the developer's mind. In this paper, we first analyze the existing classname recommendation approach for dynamically typed language. In this approach, we represent the natureor behavior of python classes in quantitative form using the embedding concept for heterogeneous graphs.Next, we use these embeddings to suggest class names. The first approach can only suggest the existing classnames. Therefore, we propose a new approach, which is based on the convolution attention model. In the newapproach, we try to generate class name as a token sequence, instead of whole class name at once. We use twovariants of the attention mechanism: simple attention and copy attention. Copy attention based model is ableto predict out-of-vocab tokens during prediction. Experimental results suggest that the convolution attentionmodel can predict accurate class name tokens.
Keyword:
Class name
Source code
Graph embeddings
Recommendation
Convolution network
Name prediction
期刊
IF:
7.5
论文数:
2.9W
被引数:
10.2W
机构
引用论文
.png){kind=icon}
Analysis and assessment of ship collision accidents using Fault Tree and Multiple Correspondence Analysis基于故障树和多重对应分析的船舶碰撞事故分析与评估
PREAI
分享收藏
分享收藏分享收藏 没有更多内容
没有更多内容