文献类型：期刊文献

英文题名：Effective approaches to combining lexical and syntactical information for code summarization

作者：Zhou, Ziyi[1];Yu, Huiqun[1,2];Fan, Guisheng[1]

机构：[1]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai, Peoples R China;[2]Shanghai Engn Res Ctr Smart Energy, Shanghai, Peoples R China

年份：2020

卷号：50

期号：12

起止页码：2313

外文期刊名：SOFTWARE-PRACTICE & EXPERIENCE

收录：;EI(收录号：20203909236388);WOS:【SCI-EXPANDED(收录号:WOS:000572222500001)】；

基金：Educational Research Fund of ECUST, Grant/Award Number: ZH1726108; National Natural Science Foundation of China, Grant/Award Numbers: 61702334, 61772200; Natural Science Foundation of Shanghai, Grant/Award Numbers: 17ZR1406900, 17ZR1429700; Shanghai Pujiang Talent Program, Grant/Award Number: 17PJ1401900

语种：英文

外文关键词：code summarization; deep learning; program comprehension

摘要：Natural language summaries of source codes are important during software development and maintenance. Recently, deep learning based models have achieved good performance on the task of automatic code summarization, which encode token sequence or abstract syntax tree (AST) of code with neural networks. However, there has been little work on the efficient combination of lexical and syntactical information of code for better summarization quality. In this paper, we propose two general and effective approaches to leveraging both types of information: a convolutional neural network that aims to better extract vector representation of AST node for downstream models; and a Switch Network that learns an adaptive weight vector to combine different code representations for summary generation. We integrate these approaches into a comprehensive code summarization model, which includes a sequential encoder for token sequence of code and a tree based encoder for its AST. We evaluate our model on a large Java dataset. The experimental results show that our model outperforms several state-of-the-art models on various metrics, and the proposed approaches contribute a lot to the improvements.