From Word Embeddings to Document Similarities for Improved Information Retrieval in Software Engineering

By Xin Ye, Hui Shen, Xiao Ma, Razvan C. Bunescu, Chang Liu. International Conference on Software Engineering, 2016

313 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/3041a9265afb2ebdb4915aa9572668bb7f32b0ef

How researchers see this paper

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

In Artificial Intelligence

In Information Systems

Citations in the corpus

Deep Learning to Find Bugs (2017, methodology): “, for API methods [29], or for terms that occur both in programs and natural language documentation [45].”
Retrieval on source code: a neural code search (2018, background): “This overall process is fairly standard in NLP and document retrieval [17], but the details for code retrieval are sometimes significantly different, as we point out where relevant.”
Context2Name: A Deep Learning-Based Approach to Infer Natural Variable Names from Usage Contexts (2017, background): “ors [25]. Beyond pure natural language texts, embeddings for code elements have been proposed, e.g., for API elements [27] and for terms that appear both in code and in natural language documentation [41]. Our embeddings differ from these approaches by serving a different purpose, namely to compress”
Are the Code Snippets What We Are Searching for? A Benchmark and an Empirical Study on Code Search with Natural-Language Queries (2020, methodology): “The DL-based methods are slow in training, but can be fast in code search too; however, YeSearch is much slower than Code-nn in code search.”
Deep Code Search (2018, methodology): “Besides code search, there have been many other information retrieval tasks in software engineering [8, 9, 16, 23, 24, 29, 51, 55, 63, 67] such as bug localization [66, 73, 80], feature localization [19], traceability links recovery [20] and community Question Answering [11].”
Learning-Based Recursive Aggregation of Abstract Syntax Trees for Code Clone Detection (2019, background): “[22], [23] or even mixed vocabularies of code tokens and natural language words [24].”
DeepBugs: a learning approach to name-based bug detection (2018, methodology): “2017], for terms that occur both in programs and natural language documentation [Ye et al. 2016], as well as a generic, AST-based representation of code [Alon et al.”
NL2Type: Inferring JavaScript Function Types from Natural Language Information (2019, background): “To the best of our knowledge, NL2Type is the ﬁrst to predict types from comments. d) Natural Language Information and Code: Code search allows developers to ﬁnd code snippets through nat-ural language queries [43]–[46].”
Neural information retrieval: at the end of the early years (2017, methodology): “Similarly for word2vec, Mitra et al. (2016) simply adopt CBOW, while others adopt Skip-gram (De Vine et al. 2014; Manotumruksa et al. 2016; Vulic and Moens 2015; Yang et al. 2016b; Ye et al. 2016; Zhou et al. 2015).”
CoSQA: 20,000+ Web Queries for Code Search and Question Answering (2021, background): “…Rah-man, 2019). deep learning based models encode query and code into vectors and utilize vector similarities as the metric to retrieve code (Sachdev et al., 2018; Ye et al., 2016; Gu et al., 2018; Cambronero et al., 2019; Yao et al., 2019; Liu et al., 2019a; Feng et al., 2020; Zhao and Sun, 2020).”