A Neural Network Architecture for Program Understanding Inspired by Human Behaviors

TL;DR

This paper introduces PGNN-EK, a neural network model inspired by human behaviors for program understanding, combining graph neural networks and external knowledge to improve code comprehension tasks.

Contribution

The paper proposes a novel model that integrates human-inspired partitioning and external knowledge for enhanced program understanding, along with a new challenging dataset.

Findings

PGNN-EK outperforms existing models on code summarization and clone detection.

The model demonstrates strong generalization on a newly released dataset.

External knowledge integration significantly improves understanding accuracy.

Abstract

Program understanding is a fundamental task in program language processing. Despite the success, existing works fail to take human behaviors as reference in understanding programs. In this paper, we consider human behaviors and propose the PGNN-EK model that consists of two main components. On the one hand, inspired by the "divide-and-conquer" reading behaviors of humans, we present a partitioning-based graph neural network model PGNN on the upgraded AST of codes. On the other hand, to characterize human behaviors of resorting to other resources to help code comprehension, we transform raw codes with external knowledge and apply pre-training techniques for information extraction. Finally, we combine the two embeddings generated from the two components to output code embeddings. We conduct extensive experiments to show the superior performance of PGNN-EK on the code summarization and code clone detection tasks. In particular, to show the generalization ability of our model, we release a new dataset that is more challenging for code clone detection and could advance the development of the community. Our codes and data are publicly available at https://github.com/RecklessRonan/PGNN-EK.