Natural Language Models for Predicting Programming Comments

By Dana Movshovitz-Attias, William W. Cohen. 2013

132 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

A Survey of Machine Learning for Big Code and Naturalness (2017, methodology): “stributed Explain code changes Maddison and Tarlow [126]∗ LM AST Context LM (AST) Distributed — Mangal et al. [127] Logic + Feedback Prob. Analysis MaxSAT Program Analysis Movshovitz-Attias and Cohen [133] Tokens Code Comments Directed GM Comment Prediction Mou et al. [132] Syntax Classiﬁcation Dist”
Learning to Mine Aligned Code and Natural Language Pairs from Stack Overflow (2018, background): “ey use to mine their data? Our work falls in the line of mining intent-snippet pairs from SO (e.g., [15, 44, 46, 48]), while there has been research on miningfromotherdatasourcessuchasAPIdocumentation[5,6,24], code comments [43], specialized sites [32], parameter/method/class names [1, 37], and deve”
Bugram: Bug detection with n-gram language models (2016, methodology): “[29] leveraged n-gram models to predict class comments for program source file documents.”
CoaCor: Code Annotation for Code Retrieval with Reinforcement Learning (2019, methodology): “Earlier works tackled the problem using template-based approaches [37, 49] and topic n-grams models [38] while the recent techniques [2, 19–22, 31] are mostly built upon deep neural networks.”
Improving Automatic Source Code Summarization via Deep Reinforcement Learning (2018, methodology): “The existing works for code summarization can be mainly categorized as rule based approaches [42], statistical language model based approaches [30] and deep learning based approaches [2, 13, 15].”
TranS^3: A Transformer-based Framework for Unifying Code Summarization and Code Search (2020, methodology): “tched in open-source projects. Similarly they also apply code clone detection techniques to •nd similar code snippets and extract comments from the similar code snippets [55]. Movshovitz-A−ias et al. [2] predicted comments from Java source •les using topic models and n-grams. Haiduc et al. [56] comb”
Using machine translation for converting Python 2 to Python 3 code (2015, methodology): “Recent works as Nguyen et al. (2013), and Movshovitz-Attias and Cohen (2013) have used n-grams language models to show the predictability of the software, and using that to generate the project documentation for the source code.”
Learning natural coding conventions (2014, background): “Formally, let L be the set of locations in x at which N ATURALIZE is able to make suggestions, and for each ‘ ∈ L , let C ‘ be the system’s set of suggestions at ‘ .”
A Neural Model for Generating Natural Language Summaries of Program Subroutines (2019, background): “Yin et al. [21], Movshovitz et al. [30], and Allamanis et al. [31] target comments of short snippets of code, a task facilitated by public datasets [32].”
On the localness of software (2014, background): “Statistical language modeling has also emerged as an approach to exploit this abundance of data [5, 21, 32, 33].”
Where should I comment my code? A dataset and model for predicting locations that need comments (2020, methodology): “Pioneering work started with rules or templates [2], before work moved to comparing 𝑛 -gram language models to LDA models for generating class comments [11].”
Can we generate shellcodes via natural language? An empirical study (2021, background): “(2015) pioneered the task of translating python code to pseudo-code while others proposed an n-gram language model to generate comments from source code (Movshovitz-Attias and Cohen 2013).”
Autofolding for Source Code Summarization (2014, methodology): “Recent work has applied language modelling [20], [21], [22], [23], [24], natural language generation [12], [25], machine translation [26], and topic modelling [27] to the text of source code”
Deep Code Comment Generation (2018, background): “Automatic code summarization approaches vary from manuallycrafted template [24, 35, 36], IR [14, 15, 43] to learning-based approaches [4, 19, 28].”
MCoNaLa: A Benchmark for Code Generation from Multiple Natural Languages (2022, background): “, 2018b), other work also attempts to exploit other data sources such as API documentation (Chatterjee et al., 2009; Movshovitz-Attias and Cohen, 2013; Xu et al., 2020), code comments (Wong et al.”