Web API evolution patterns: A usage-driven approach

As the use of Application Programming Interfaces (APIs) is increasingly growing, their evolution becomes more challenging in terms service provided according to consumers’ needs. In this paper, we address role needs WAPIs and introduce a process mining pattern-based method support providers by analyzing understanding behavior, imprinted WAPI usage logs. We take position that WAPIs’ should be mainly usage-based, i.e., way consumers them one main drivers changes. start characterizing structural relationships between endpoints, next, summarize these into set behavioral patterns (i.e., whose occurrences indicate specific behavior like repetitive or consecutive calls), can potentially imply need for changes (e.g., creating new parameters merging endpoints). analyze logs extract several metrics endpoints relationships, then detect patterns. apply our two real-world from different domains, education, health, respectively Barcelona School Informatics at Polytechnic University Catalonia (Facultat d’Informàtica de Barcelona, FIB, UPC), District Health Information Software 2 (DHIS2) WAPI. The feedback proved effectiveness detected confirmed promising potential approach.