Two Algorithms for Architecture Pattern Recognition

P.Y. Piñero, M. Cuesta, R. Lazo Ochoa (Cuba), and W. NonPakdee (Thailand)


Architectural Patterns, Software Engineer and Artificial Intelligence


The architectural pattern recognition is an interesting research area for programmers and software engineers. The appropriate selection and application of architectural pattern during the software development guarantee flexibility and high quality of software. However, usually, it is difficult to obtain a robust design with features such as: reusability, flexibility, clearly and high performance. In this paper, the authors present a brief overview of architectural patterns and discuss different trends in the architectural patterns design such as: Gang-Of-Four Book and J2EE patterns. This work aims to present a new model for architectural pattern recognition based on artificial intelligence techniques and pattern recognitions strategies. Two new algorithms are discussed in details. This model constitutes a useful solution and should be used during the architecture design faces regardless of a specific pattern catalogue. However, in this paper, the authors present experimental results based on Gang-Of Four pattern catalogue only. This pattern recognition model consists of making decision tool for programmers and software engineers. The presented algorithms do not aim to substitute the human experience. Moreover, the algorithms results and suggestions design should always be revised by human experts.

Important Links:

Go Back