Research Landscape of Patterns in Software Engineering: Taxonomy, State-of-the-Art, and Future Directions

Software system development is a difficult, error-prone, and time-consuming task. Pattern-based development has been successfully used to produce high-quality software to address these concerns. Patterns actually document expert knowledge on how to solve recurring problems. As a result, they provide a valuable foundation for experience sharing and reuse. Pattern-based strategies have become increasingly popular in recent years. In the literature, several secondary studies dealing with this concept have been proposed. Nevertheless, they either focus on specific sub-fields (domain-specific or intention-specific) or give only quantitative analysis, with no intention of classifying and synthesizing the state-of-the-art of pattern-based approaches. The objective of this work is to provide a comprehensive survey of the literature reporting outcomes from various research communities and domains. We present a taxonomy that serves as a basis for categorizing and analyzing 134 papers from various scientific areas. The findings may serve as a guide for novice researchers in this field on the one hand, and may assist the research community in consolidating their research efforts through the description of future directions on the other.