检索结果-内蒙古大学图书馆

Towards Retrieval-Based Neural code summarization: A Meta-Learning Approach

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 2023年第4期49卷 3008-3031页

作者： Zhou, Ziyi Yu, Huiqun Fan, Guisheng Huang, Zijie Yang, Kang East China Univ Sci & Technol Comp Sci Shanghai Peoples R China Nanjing Univ Comp Sci Nanjing Peoples R China East China Univ Sci & Technol ECUST Comp Sci Shanghai Peoples R China Shanghai Jiao Tong Univ Comp Sci Shanghai Peoples R China ECUST Dept Comp Sci & Engn Comp Sci Shanghai Peoples R China Florida Int Univ Sch Comp Sci Miami FL 33199 USA Anhui Univ Technol Comp Sci Maanshan Anhui Peoples R China East China Univ Sci & Technol ECUST Comp Sci Shanghai Peoples R China East China Univ Sci & Technol Comp Sci Shanghai Peoples R China East China Univ Sci & Technol Dept Comp Sci & Engn Shanghai Peoples R China Shanghai Normal Univ Comp Sci Shanghai Peoples R China East China Univ Sci & Technol Comp Sci Shanghai Peoples R China

code summarization aims to generate code summaries automatically, and has attracted a lot of research interest lately. Recent approaches to it commonly adopt neural machine translation techniques, which train a Seq2Seq model on a large corpus and assume it could work on various new code snippets. However, codes are highly varied in practice due to different domains, businesses or programming styles. Therefore, it is challenging to learn such a variety of patterns into a single model. In this paper, we propose a brand-new framework for code summarization based on meta-learning and code retrieval, named MLCS to tackle this issue. In this framework, the summarization of each target code is formalized as a few-shot learning task, where its similar examples are used as training data and the testing example is itself. We retrieve examples similar to the target code in a rank-and-filter manner. Given a neural code summarizer, we optimize it into a meta-learner via Model-Agnostic Meta-Learning (MAML). During inference, the meta-learner first adapts to the retrieved examples and yields an exclusive model for the target code, and then generates its summary. Extensive experiments on real-world datasets show: (1) Utilizing MLCS, a standard Seq2Seq model is able to outperform previous state-of-the-art approaches, including both neural models and retrieval-based neural models;(2) MLCS can flexibly adapt to existing neural code summarizers without modifying their architecture, and could significantly improve their performance with the relative gain of up to 112.7% on BLEU-4, 23.2% on ROUGE-L, and 31.5% on METEOR;(3) Compared to the existing retrieval-based neural approaches, MLCS can better leverage multiple similar examples, and shows better generalization ability on different retrievers, unseen retrieval corpus and low-frequency words.

关键词： code summarization deep learning information retrieval meta-learning

来源：评论

学校读者我要写书评

暂无评论

A Tale of Two Comprehensions? Analyzing Student Programmer Attention during code summarization

引用

ACM TRANSACTIONS ON SOFTWARE ENGINEERING AND METHODOLOGY 2024年第7期33卷 1-37页

作者： Karas, Zachary Bansal, Aakash Zhang, Yifan Li, Toby Mcmillan, Collin Huang, Yu Vanderbilt Univ Comp Sci Nashville TN 37235 USA Univ Notre Dame Comp Sci & Engn South Bend IN USA Univ Notre Dame Dept Comp Sci & Engn South Bend IN USA

code summarization is the task of creating short, natural language descriptions of source code. It is an important part of code comprehension and a powerful method of documentation. Previous work has made progress in identifying where programmers focus in code as they write their own summaries (i.e., Writing). However, there is currently a gap in studying programmers' attention as they read code with pre-written summaries (i.e., Reading). As a result, it is currently unknown how these two forms of code comprehension compare: Reading and Writing. Also, there is a limited understanding of programmer attention with respect to program semantics. We address these shortcomings with a human eye-tracking study (n = 27) comparing Reading and Writing. We examined programmers' attention with respect to fine-grained program semantics, including their attention sequences (i.e., scan paths). We find distinctions in programmer attention across the comprehension tasks, similarities in reading patterns between them, and differences mediated by demographic factors. This can help guide code comprehension in both computer science education and automated code summarization. Furthermore, we mapped programmers' gaze data onto the Abstract Syntax Tree to explore another representation of human attention. We find that visual behavior on this structure is not always consistent with that on source code.

关键词： Additional Key Words and Phrases Cognitive science code summarization eye-tracking expertise code comprehension

来源：评论

学校读者我要写书评

暂无评论

Effective approaches to combining lexical and syntactical information for code summarization

引用

SOFTWARE-PRACTICE & EXPERIENCE 2020年第12期50卷 2313-2336页

作者： Zhou, Ziyi Yu, Huiqun Fan, Guisheng East China Univ Sci & Technol Dept Comp Sci & Engn Shanghai Peoples R China Shanghai Engn Res Ctr Smart Energy Shanghai Peoples R China

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.

关键词： code summarization deep learning program comprehension

来源：评论

学校读者我要写书评

暂无评论

Adversarial training and ensemble learning for automatic code summarization

引用

NEURAL COMPUTING & APPLICATIONS 2021年第19期33卷 12571-12589页

作者： Zhou, Ziyi Yu, Huiqun Fan, Guisheng East China Univ Sci & Technol Dept Comp Sci & Engn Shanghai Peoples R China Shanghai Engn Res Ctr Smart Energy Shanghai Peoples R China

Natural language summaries of codes are important during software development and maintenance. Recently, deep learning-based models have achieved good performance on automatic code summarization, which encode token sequence or abstract syntax tree (AST) of code with neural networks. However, almost all of these models are trained using maximum likelihood estimation, which do not guarantee the quality of generated summaries. Moreover, existing models that benefit from multiple encoders lack a fined-grained selection between different encoders, and the encoders may be insufficiently optimized. To address these issues and generate better code summaries, we propose a novel code summarization framework based on adversarial training and ensemble learning. It includes two separately trained encoder-decoder models, one for source code sequence and the other for its AST. Here, an efficient approach to obtain AST node sequence is introduced. We train our models via adversarial training, where each model is guided by a well-designed discriminator that learns to evaluate its outputs. During inference, a module named mixture network is introduced to compute an adaptive combination weight of the models' outputs. We evaluate our framework on a large Java corpus and compare it to several state-of-the-art models. Experimental results show that our approach outperforms the best baseline by 22.6% on BLEU-4, 5.7% on ROUGE-L and 7.6% on METEOR.

关键词： code summarization Program comprehension Deep learning

来源：评论

学校读者我要写书评

暂无评论

Why My code summarization Model Does NotWork: code Comment Improvement with Category Prediction

引用

ACM TRANSACTIONS ON SOFTWARE ENGINEERING AND METHODOLOGY 2021年第2期30卷 p1-29页

作者： Chen, Qiuyuan Xia, Xin Hu, Han Lo, David Li, Shanping Zhejiang Univ Coll Comp Sci & Technol 38 Zheda Rd Hangzhou 310000 Zhejiang Peoples R China Monash Univ Fac Informat Technol Wellington Rd Melbourne Vic Australia Singapore Management Univ Sch Informat Syst 81 Victoria St Singapore 188065 Singapore

code summarization aims at generating a code comment given a block of source code and it is normally performed by training machine learning algorithms on existing code block-comment pairs. code comments in practice have different intentions. For example, some code comments might explain how the methods work, while others explain why some methods are written. Previous works have shown that a relationship exists between a code block and the category of a comment associated with it. In this article, we aim to investigate to which extent we can exploit this relationship to improve code summarization performance. We first classify comments into six intention categories and manually label 20,000 code-comment pairs. These categories include "what," "why," "how-to-use," "how-it-is-done," "property," and "others." Based on this dataset, we conduct an experiment to investigate the performance of different state-of-the-art code summarization approaches on the categories. We find that the performance of different code summarization approaches varies substantially across the categories. Moreover, the category for which a code summarization model performs the best is different for the different models. In particular, no models perform the best for "why" and "property" comments among the six categories. We design a composite approach to demonstrate that comment category prediction can boost code summarization to reach better results. The approach leverages classified code-category labeled data to train a classifier to infer categories. Then it selects the most suitable models for inferred categories and outputs the composite results. Our composite approach outperforms other approaches that do not consider comment categories and obtains a relative improvement of 8.57% and 16.34% in terms of ROUGE-L and BLEU-4 score, respectively.

关键词： code summarization code comment comment classification

来源：评论

学校读者我要写书评

暂无评论

Reinforcement-Learning-Guided Source code summarization Using Hierarchical Attention

引用

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 2022年第1期48卷 102-119页

作者： Wang, Wenhua Zhang, Yuqun Sui, Yulei Wan, Yao Zhao, Zhou Wu, Jian Yu, Philip S. Xu, Guandong Southern Univ Sci & Technol Dept Comp Sci & Engn Shenzhen 518055 Guangdong Peoples R China Univ Technol Sch Comp Sci Sydney NSW 2008 Australia Zhejiang Univ Hangzhou 310027 Zhejiang Peoples R China Univ Illinois Chicago IL 60607 USA

code summarization (aka comment generation) provides a high-level natural language description of the function performed by code, which can benefit the software maintenance, code categorization and retrieval. To the best of our knowledge, the state-of-the-art approaches follow an encoder-decoder framework which encodes source code into a hidden space and later decodes it into a natural language space. Such approaches suffer from the following drawbacks: (a) they are mainly input by representing code as a sequence of tokens while ignoring code hierarchy;(b) most of the encoders only input simple features (e.g., tokens) while ignoring the features that can help capture the correlations between comments and code;(c) the decoders are typically trained to predict subsequent words by maximizing the likelihood of subsequent ground truth words, while in real world, they are excepted to generate the entire word sequence from scratch. As a result, such drawbacks lead to inferior and inconsistent comment generation accuracy. To address the above limitations, this paper presents a new code summarization approach using hierarchical attention network by incorporating multiple code features, including type-augmented abstract syntax trees and program control flows. Such features, along with plain code sequences, are injected into a deep reinforcement learning (DRL) framework (e.g., actor-critic network) for comment generation. Our approach assigns weights (pays "attention") to tokens and statements when constructing the code representation to reflect the hierarchical code structure under different contexts regarding code features (e.g., control flows and abstract syntax trees). Our reinforcement learning mechanism further strengthens the prediction results through the actor network and the critic network, where the actor network provides the confidence of predicting subsequent words based on the current state, and the critic network computes the reward values of all the possible ex

关键词： code summarization hierarchical attention reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

CLG-Trans: Contrastive learning for code summarization via graph attention-based transformer

引用

SCIENCE OF COMPUTER PROGRAMMING 2023年第1期226卷

作者： Zeng, Jianwei He, Yutong Zhang, Tao Xu, Zhou Han, Qiang Macau Univ Sci & Technol Sch Comp Sci & Engn Taipa Macao Peoples R China Chongqing Univ Sch Big Data & Software Engn Chongqing Peoples R China North Minzu Univ Sch Comp Sci & Engn Yinchuan Ningxia Peoples R China

Automated code summarization is the task of automatically generating natural language descriptions of source code, which is an important research topic in the software engineering field. Many methods in recent studies were based on deep learning techniques, which effectively improve the performance of code summarization. Most of the existing code summarization methods use different kinds of neural networks to learn source code information. Some methods use graph neural network (GNN) to represent abstract syntax tree (AST) and fuse the structural information of source code. However, these methods still have two important issues: 1) they cannot solve the Out-Of-Vocabulary (OOV) problem effectively;2) the structural information of source code they can capture is limited. In order to address the above-mentioned challenges, we propose a novel automated code summarization model named CLG-Trans in this work. This model uses the Byte Pair Encoding (BPE) algorithm and pointer-generator network to tackle the OOV problem. Then it utilizes the fusion of contrastive learning strategy and dynamic graph attention mechanism to effectively capture rich structure information of source code sequences. Experimental results on Funcom dataset show that CLG-Trans outperforms seven state-ofthe-art models (i.e., Hybrid-DRL, Ast-Attendgru, Transformer, codeGnn, Rencos, codeBERT and SIT) by averagely increasing 19.48% and 13.17% on BLEU scores and ROUGUE-L score, respectively. In addition, CLG-Trans achieves an improvement of 16.14% and 4.70% in BLEU scores and ROUGE-L score compared with our previously proposed model DG-Trans. (c) 2023 Elsevier B.V. All rights reserved.

关键词： code summarization Graph neural networks Dynamic graph attention Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Deep Is Better? An Empirical Comparison of Information Retrieval and Deep Learning Approaches to code summarization

引用

ACM TRANSACTIONS ON SOFTWARE ENGINEERING AND METHODOLOGY 2024年第3期33卷 1-37页

作者： Zhu, Tingwei Li, Zhong Pan, Minxue Shi, Chaoxuan Zhang, Tian Pei, Yu Li, Xuandong Nanjing Univ State Key Lab Novel Software Technol 163 Xianlin Ave Nanjing 210023 Jiangsu Peoples R China Nanjing Univ Dept Comp Sci & Technol 163 Xianlin Ave Nanjing 210023 Jiangsu Peoples R China Nanjing Univ State Key Lab Novel Software Technol Software Inst 22 Hankou Rd Nanjing 210093 Jiangsu Peoples R China Hong Kong Polytech Univ Dept Comp Hong Kong Peoples R China

code summarization aims to generate short functional descriptions for source code to facilitate code comprehension. While Information Retrieval (IR) approaches that leverage similar code snippets and corresponding summaries have led the early research, Deep Learning (DL) approaches that use neural models to capture statistical properties between code and summaries are now mainstream. Although some preliminary studies suggest that IR approaches are more effective in some cases, it is currently unclear how effective the existing approaches can be in general, where and why IR/DL approaches perform better, and whether the integration of IR and DL can achieve better performance. Consequently, there is an urgent need for a comprehensive study of the IR and DL code summarization approaches to provide guidance for future development in this area. This article presents the first large-scale empirical study of 18 IR, DL, and hybrid code summarization approaches on five benchmark datasets. We extensively compare different types of approaches using automatic metrics, we conduct quantitative and qualitative analyses of where and why IR and DL approaches perform better, respectively, and we also study hybrid approaches for assessing the effectiveness of integrating IR and DL. The study shows that the performance of IR approaches should not be underestimated, that while DL models perform better in predicting tokens from method signatures and capturing structural similarities in code, simple IR approaches tend to perform better in the presence of code with high similarity or long reference summaries, and that existing hybrid approaches do not perform as well as individual approaches in their respective areas of strength. Based on our findings, we discuss future research directions for better code summarization.

关键词： code summarization deep learning information retrieval

来源：评论

学校读者我要写书评

暂无评论

CoSS: Leveraging Statement Semantics for code summarization

引用

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 2023年第6期49卷 3472-3486页

作者： Shi, Chaochen Cai, Borui Zhao, Yao Gao, Longxiang Sood, Keshav Xiang, Yong Deakin Univ Sch Informat Technol Geelong Vic 3220 Australia Qilu Univ Technol Shandong Acad Sci Jinan 250103 Shandong Peoples R China

Automated code summarization tools allow generating descriptions for code snippets in natural language, which benefits software development and maintenance. Recent studies demonstrate that the quality of generated summaries can be improved by using additional code representations beyond token sequences. The majority of contemporary approaches mainly focus on extracting code syntactic and structural information from abstract syntax trees (ASTs). However, from the view of macro-structures, it is challenging to identify and capture semantically meaningful features due to fine-grained syntactic nodes involved in ASTs. To fill this gap, we investigate how to learn more code semantics and control flow features from the perspective of code statements. Accordingly, we propose a novel model entitled CoSS for code summarization. CoSS adopts a Transformer-based encoder and a graph attention network-based encoder to capture token-level and statement-level semantics from code token sequence and control flow graph, respectively. Then, after receiving two-level embeddings from encoders, a joint decoder with a multi-head attention mechanism predicts output sequences verbatim. Performance evaluations on Java, Python, and Solidity datasets validate that CoSS outperforms nine state-of-the-art (SOTA) neural code summarization models in effectiveness and is competitive in execution efficiency. Further, the ablation study reveals the contribution of each model component.

关键词： Attention mechanism code summarization control flow graph graph neural network

来源：评论

学校读者我要写书评

暂无评论

ClassSum: a deep learning model for class-level code summarization

引用

NEURAL COMPUTING & APPLICATIONS 2023年第4期35卷 3373-3393页

作者： Li, Mingchen Yu, Huiqun Fan, Guisheng Zhou, Ziyi Huang, Jiawen East China Univ Sci & Technol Dept Comp Sci & Engn Shanghai Peoples R China Shanghai Engn Res Ctr Smart Energy Shanghai Peoples R China Shanghai Comp Software Tech Dev Ctr Shanghai Key Lab Comp Software Evaluating & Testi Shanghai Peoples R China

code summaries are clear and concise natural language descriptions of program entities. Meaningful code summaries assist developers in better understanding. code summarization refers to the task of generating a natural language summary from a code snippet. Most researches on code summarization focus on automatically generating summaries for methods or functions. However, in an object-oriented language such as Java, class is the basic programming unit rather than method. To fill this gap, in this paper, we investigate how to generate summaries for Java classes utilizing deep learning-based approaches. We propose a novel encoder-decoder model called ClassSum to generate functionality descriptions for Java classes and build a dataset containing 172,639 pairs from 3185 repositories hosted on Github. Since the code of class is much longer and more complicated, encoding a whole class via neural network is more challenging than encoding a method. On the other hand, the content within a class may be incomplete. To overcome this difficulty, we reduce the code of a class by only keeping its key elements, namely class signatures, method signatures and attribute names. To utilize both lexical and structural information of code, our model takes token sequence and abstract syntax tree of the reduced class content as inputs. ClassSum and five baselines (designed for method-level code summarization) are evaluated on our dataset. Experiment results show that summaries generated by ClassSum are more accurate and readable than those generated by baselines. Our dataset is available at https://***/classsum/ClassSum.

关键词： Program comprehension code summarization Class documentation Deep learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：