Predicting Web Service Maintainability Via Object-Oriented Metrics: a Statistics-Based Approach

Abstract

The Service-Oriented Computing paradigm enables the construction of distributed systems by assembling loosely coupled pieces of software called services, which have clear interfaces to their functionalities. Service interface descriptions have many aspects, such as complexity and quality, all of which can be measured. This paper presents empirical evidence showing that services interfaces maintainability can be predicted by applying traditional software metrics in service implementations. A total of 11 source code level metrics and 5 service interface metrics have been statistically correlated using 154 real world services. © 2012 Springer-Verlag.