Advanced Detection of Source Code Clones via an Ensemble of Unsupervised Similarity Measures

TL;DR

The paper tackles code similarity and clone detection under limited labeled data by proposing an ensemble of unsupervised similarity measures. It formalizes the task as a function $f: S \times S \rightarrow [0,1]$ and builds a weighted ensemble $E(s_a,s_b) = \sum_i α_i m_i(s_a,s_b)$ learned via bagging and boosting. Empirically, ensembles surpass single unsupervised measures and reach competitive performance with transformer-based models on small datasets, while offering better interpretability and lower training carbon footprint. The work also provides open-source code and discusses practical deployment for tasks like clone detection and plagiarism assessment.

Abstract

The capability of accurately determining code similarity is crucial in many tasks related to software development. For example, it might be essential to identify code duplicates for performing software maintenance. This research introduces a novel ensemble learning approach for code similarity assessment, combining the strengths of multiple unsupervised similarity measures. The key idea is that the strengths of a diverse set of similarity measures can complement each other and mitigate individual weaknesses, leading to improved performance. Preliminary results show that while Transformers-based CodeBERT and its variant GraphCodeBERT are undoubtedly the best option in the presence of abundant training data, in the case of specific small datasets (up to 500 samples), our ensemble achieves similar results, without prejudice to the interpretability of the resulting solution, and with a much lower associated carbon footprint due to training. The source code of this novel approach can be downloaded from https://github.com/jorge-martinez-gil/ensemble-codesim.