City Research Online

Abstract

In software development, the identification and utilization of design patterns play a crucial role in enhancing code quality, reusability, and maintainability. Design patterns are reusable solutions to common problems encountered in software design and development. Detecting design patterns within source code can aid developers in recognizing recurring patterns of design decisions, promoting best practices, and improving software scalability and readability. Our proposed method, RISC-dpi (Role Identification and Sequence Classification for Design Patterns Identification), offers an automated approach to extracting call sequences from source code and identifying sequences that exhibit similar behavior to target design patterns. RISC-dpi transforms the call sequences of the code to a graph then uses hidden Markov Model (HMM) to detect sequences that have a similar behavior to the target design pattern. RISC-dpi consists of two phases. In the first phase, the role of each class instance of sequence is identified using classification and in the second phase, it is determined whether each extracted sequence is a design pattern or not. To evaluate RISC-dpi, we applied it on the P-MARt dataset and compared its results with the ones obtained by different methods. Experimental results reveal that RISC-dpi can detect all kinds of design patterns with high performance. We also obtained better recall, precision and F-measure compared to the previous learning-based methods.

Publication Type:	Article
Additional Information:	This version of the article has been accepted for publication, after peer review and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s42979-026-04838-4
Publisher Keywords:	Design pattern detection, Sequence classification, Graph transformation, and Machine learning
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	School of Science & Technology School of Science & Technology > Department of Computer Science
SWORD Depositor:	Symplectic Administrator

Export

Downloads