A Fuzzy Approach for Integrated Measure of Object Oriented Software Complexity

Abstract

Many methods can be used to measure the complexity of a Object
Oriented software, but none o f them takes care of all the
characteristics o f the software. Various metrics used in
measurement o f OO software complexity are Weighted Methods
per Class (WMC), Response for a Class (RFC), Lack of
Cohesion of Methods (LCOM), Coupling Between Objects
(CBO), Depth Inheritance Tree (DIT) and Number of Children
(NOC), but none o f them is alone sufficient to give an overall
reflection of software complexity. Different metrics try to
measure different characteristics o f the software such as
efficiency, complexity, understandability, reusability, testability
and maintainability. Depending upon the characteristics o f the
software to be measured a subset o f thkse metrics should be used.
This paper proposes a fuzzy model for complexity measurement
that integrates the effect of subset o f these metrics i.e. WMC,
RFC, CBO and DIT. The paper is about an approach to combine
OO software metric values into a single overall value that can be
used to rank classes according to their'complexity. The
approach uses fuzzy techniques and concepts, vizfuzzification of
crisp metric values, inference and aggregation, defuzzification
of fuzzy output etc.