Home // ICSEA 2023, The Eighteenth International Conference on Software Engineering Advances // View article

Design Pattern Detection in Code: A Hybrid Approach Utilizing a Bayesian Network, Machine Learning with Graph Embeddings, and Micropattern Rules

Authors:
Roy Oberhauser
Sandro Moser

Keywords: software design pattern detection; machine learning; artificial neural networks; graph embeddings; rule-based expert system; Bayesian networks; software engineering

Abstract:
Software design patterns and the abstractions they offer can support developers and maintainers with program code comprehension. Yet manually-created pattern documentation within code or code-related assets, such as documents or models, can be unreliable, incomplete, and labor-intensive. While various Design Pattern Detection (DPD) techniques have been proposed, industrial adoption of automated DPD remains limited. This paper contributes a hybrid DPD solution approach that leverages a Bayesian network integrating developer expertise via rule-based micropatterns with our machine learning subsystem that utilizes graph embeddings. The prototype shows its feasibility, and the evaluation using three design patterns shows its potential for detecting both design patterns and variations.

Pages: 122 to 129

Copyright: Copyright (c) IARIA, 2023

Publication date: November 13, 2023

Published in: conference

ISSN: 2308-4235

ISBN: 978-1-68558-098-8

Location: Valencia, Spain

Dates: from November 13, 2023 to November 17, 2023