Improving web service interfaces modularity using multi-objective optimization

How to Evaluate Solutions in Pareto-Based Search-Based Software Engineering: A Critical Review and Methodological Guidance

“…We used the citation information from Google Scholar as a single metric to (partially) assess the impact of a study 6 . In particular, we follow a pragmatic strategy that: a study has 5 citations per year from its year of publication is counted in, e.g., a 2010 study would expect to have at least 45 citations 7 . The only exception is for the work published in the year of writing this article (i.e., 2019), we consider those that were published for shorter than 6 months and have been cited by more than once, together with the pre-press ones that have not yet been given an issue number regardless of their citation counts.…”

“…Many Pareto-based SBSE studies evaluate solution sets by DOE -statistical objective values in the obtained solution set(s). For example, as it can be seen in Table 4, the mean objective value was considered in [6], [7], [9], [102], [105], [107], [114], [118], [128], [129]; the median value in [6], [37], [58]; the best value in [9], [21], [30], [65], [82], [88], [100], [108], [134], [135], [141], [142]; the worst value in [9], [134]; the statistical significance of the differences between distinct solution sets' objective values in [1], [76], [123], [137]. Such DOE measures need to be used in line with the DM's preference.…”

“…An example that preferring knee points while using the indicator IGD (as well as GD and CI) can lead to misleading results. Consider two solution sets (A and B) for a bi-objective minimization scenario, where A = {(2, 6), (9, 2)} are the two knee points of the Pareto front, and B = {(1, 10),(7,5), (12, 1.5)} are three well-distributed non-knee points on the Pareto front. Apparently, if the DM prefers knee points then solutions in A will certainly be selected.…”

Static Analysis for Improved Modularity of Procedural Web Application Programming Interfaces

“…Both approaches are based on traditional algorithms of greedy search and graph partitioning to address this problem. Boukharata et al [41] extended the work of Ouni et al [40] to determine sequential similarity (input/output dependency), communication similarity (message types) and semantic similarity (data types related to domain concepts). These extracted similarity measures are used through multi-objective optimization (using NSGA-II), to find optimal modularization of operations, reaching the best tradeoff between minimizing coupling, maximizing cohesion, and minimizing the interfaces modifications.…”

A URI parsing technique and algorithm for anti-pattern detection in RESTful Web services

“…One of the typical poor RESTful WS design practices is using unrelated names for resources or parameters and then group them in service (Aghajani et al , 2018; Palma et al , 2017). Consequently, such services are difficult to understand and reuse by other users (or programmes) and often result in underused services (Boukharata et al , 2019; Ouni et al , 2016). Indeed, to guarantee the successful utilisation of the services, users (or programmes) must understand the functions of the services before they use them (Ouni et al , 2016).…”