Reinforcement-Learning-Guided Source Code Summarization Using Hierarchical Attention

Wang, Wenhua; Zhang, Yuqun; Sui, Yulei; Wan, Yao; Zhao, Zhou; Wu, Jian; Yu, Philip S.; Xu, Guandong

doi:10.1109/tse.2020.2979701

Cited by 75 publications

(38 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our proposed technique is a deep model based on RNN and comprised of pyramids in which the outcome of the lower pyramids turns into the inputs to the higher pyramids. HANet [39] focuses on the document level classification that a document has K sentences, and each sentence contains T i words, where wit with t integral [1, T ] represents the words in the ith sentence. HANet structural design is explained in Fig.…”

Section: Classificationmentioning

confidence: 99%

A deep learning method DCWR with HANet for stock market prediction using news articles

Albahli

Awan

Nazir

et al. 2022

Complex Intell. Syst.

View full text Add to dashboard Cite

Stock movement prediction is a challenging problem to analyze in both academic and financial research areas. The advancement of deep learning (DL) techniques has grasped the attention of researchers to employ them for predicting the stock market’s future trends. Few frameworks can understand the financial terms in literature, and the volatile nature of stock markets further complicates this process. This paper has tried to overcome the existing challenges by introducing a DL-based framework using financial news articles to forecast the stock market. After performing preprocessing step, the deep contextualized word representation (DCWR) approach is applied for feature extraction. In the next step, the independent component analysis (ICA) method is used for feature reduction. Finally, the resultant features train the hierarchical attention networks (HANet) classifier to predict the stock movements. The proposed scheme is evaluated over the 7 years of data from a publicly available dataset gathered from the Reuter’s website and attained an average prediction accuracy of 92.5% which shows our framework’s robustness.

show abstract

Section: Classificationmentioning

confidence: 99%

A deep learning method DCWR with HANet for stock market prediction using news articles

Albahli

Awan

Nazir

et al. 2022

Complex Intell. Syst.

View full text Add to dashboard Cite

show abstract

“…Scanner sc = new Scanner (System.in); int A = sc.nextInt(); int B = sc.nextInt(); int C = sc.nextInt(); if( C <= A + B ) System.out.println (" Yes "); else System.out.println (" No "); Syntax n = int (input ()) if n == 12 : print ( 1 ) else : print (n + 1) Reference A , B , C = map (int ,input ().split ()) if A + B < C : print (" No ") else : print (" Yes ") Generated A , B = map (int ,input ().split ()) if A == B : print (" YES ") else : print (" NO ") et al., 2018;Wang et al, 2020b;Wan et al, 2019;Hua et al, 2021) propose to integrate the semantics of code from different views (e.g., the tokens, AST and control-flow graph) into a hybrid feature space, and put forward a hybrid representation approach, for the task of code summarization, code search and code clone detection. As for graph-based representations several works resort to parse the program into a graph (e.g., augmented AST, controlflow graph, and data-flow graph) (Li et al, 2015;LeClair et al, 2020;Wan et al, 2019;Sui et al, 2020;.…”

Section: Semanticmentioning

confidence: 99%

Disentangled Code Representation Learning for Multiple Programming Languages

Zhang¹,

Hong²,

Yin³

et al. 2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

Self Cite

View full text Add to dashboard Cite

Developing effective distributed representations of source code is fundamental yet challenging for many software engineering tasks such as code clone detection, code search, code translation and transformation. However, current code embedding approaches that represent the semantic and syntax of code in a mixed way are less interpretable and the resulting embedding can not be easily generalized across programming languages. In this paper, we propose a disentangled code representation learning approach to separate the semantic from the syntax of source code under a multi-programming-language setting, obtaining better interpretability and generalizability. Specially, we design three losses dedicated to the characteristics of source code to enforce the disentanglement effectively. We conduct comprehensive experiments on a real-world dataset composed of programming exercises implemented by multiple solutions that are semantically identical but grammatically distinguished. The experimental results validate the superiority of our proposed disentangled code representation, compared to several baselines, across three types of downstream tasks, i.e., code clone detection, code translation, and code-to-code search.

show abstract

“…They input an AST structure and sequential content of code segments into a deep reinforcement learning framework (i.e., actor-critic network). Then, Wang et al [20] extended the method Hybrid-DRL. They used a hierarchical attention network by considering multiple code features, such as type-augmented ASTs and program control flows.…”

Section: Related Workmentioning

confidence: 99%

“…In our empirical study, we choose code corpus 5 gathered by Hu et al [18] as our empirical subjects, since this code corpus have been widely used in previous studies for code comment generation [18] [11][19] [20][21] [22].…”

Section: A Code Corpusmentioning

confidence: 99%

See 1 more Smart Citation

ComFormer: Code Comment Generation via Transformer and Fusion Method-based Hybrid Code Representation

Yang¹,

Chen²,

Cao³

et al. 2021

Preprint

View full text Add to dashboard Cite

Developers often write low-quality code comments due to the lack of programming experience, which can reduce the efficiency of developers' program comprehension. Therefore, developers hope that code comment generation tools can be developed to illustrate the functionality and purpose of the code. Recently, researchers mainly model this problem as the neural machine translation problem and tend to use deep learning-based methods. In this study, we propose a novel method ComFormer based on Transformer and fusion method-based hybrid code presentation. Moreover, to alleviate OOV (out-of-vocabulary) problem and speed up model training, we further utilize the Byte-BPE algorithm to split identifiers and Sim SBT method to perform AST Traversal. We compare ComFormer with seven state-of-the-art baselines from code comment generation and neural machine translation domains. Comparison results show the competitiveness of ComFormer in terms of three performance measures. Moreover, we perform a human study to verify that ComFormer can generate high-quality comments.

show abstract

Reinforcement-Learning-Guided Source Code Summarization Using Hierarchical Attention

Cited by 75 publications

References 67 publications

A deep learning method DCWR with HANet for stock market prediction using news articles

A deep learning method DCWR with HANet for stock market prediction using news articles

Disentangled Code Representation Learning for Multiple Programming Languages

ComFormer: Code Comment Generation via Transformer and Fusion Method-based Hybrid Code Representation

Contact Info

Product

Resources

About