Write a Blog >>
CC 2021
Tue 2 - Wed 3 March 2021 Online Conference
Wed 3 Mar 2021 12:30 - 12:45 at CC Virtual Room - Natural & Source Language Analysis Chair(s): Zhijia Zhao

This paper presents Deepsy, a Natural Language-based synthesizer to assist source code analysis. It takes English descriptions of to-be-found code patterns as its inputs, and automatically produces ASTMatcher expressions that are directly usable by LLVM/Clang to materialize intended code analysis. The code analysis domain features profuse complexities in data types and operations, which make it elusive for prior rule-based synthesizers to tackle. On the other hand, machine learning-based solutions are neither applicable due to the scarcity of well labeled examples. This paper presents how Deepsy addresses the challenges by leveraging deep Natural Language Processing (NLP) and creating a new technique named dependency tree-based co-evolvement. Deepsy features an effective design that seamlessly integrates Natural Language dependency analysis into code analysis and meanwhile synergizes it with type-based narrowing and domain-specific guidance. Deepsy achieves over 70.0% expression-level accuracy and 85.1% individual API-level accuracy, significantly outperforming previous solutions.

Wed 3 Mar

Displayed time zone: Eastern Time (US & Canada) change

12:30 - 13:00
Natural & Source Language AnalysisCC Research Papers at CC Virtual Room
Chair(s): Zhijia Zhao UC Riverside
Deep NLP-Based Co-evolvement for Synthesizing Code Analysis from Natural Language
CC Research Papers
Zifan Nan North Carolina State University, Hui Guan University of Massachusetts at Amherst, Xipeng Shen North Carolina State University, Chunhua Liao Lawrence Livermore National Laboratory
Resolvable Ambiguity: Principled Resolution of Syntactically Ambiguous ProgramsArtifacts Evaluated – Reusable v1.1Results Reproduced v1.1Artifacts Available v1.1
CC Research Papers