Towards a Valid Metric for Class Cohesion at Design Level

摘要

In object oriented paradigm, cohesion of a class refers to the degree to which members of the class are interrelated. Design level class cohesion metrics are based on the assumption that if all the methods of a class have access to similar parameter types then they all process closely related information. A class with a large number of parameter types common in its methods is more cohesive than a class with less number of parameter types common in its methods. In this paper, we review the design level class cohesion metrics with a special focus on metrics which use similarity of parameter types of methods of a class as the basis of its cohesiveness. Keeping in mind the anomalies in the definitions of the existing metrics, a variant of the NHD metric is introduced. It is named NHD Modified (NHDM). The metric is analyzed with the mathematical properties of cohesion metrics as proposed in research literature. An automated metric collection tool is used to collect the metric data from an open source software program. Statistical analysis of the data shows that NHDM metric takes the lowest average value in this group of four metrics. It may be due to the fact that NHDM, unlike other metrics, does not give any false positivesegatives.