Comparative Analysis of Software Complexity of Searching Algorithms Using Code Based Metrics
Olabiyisi S.O., Omidiora E.O, Sotonwa K. A · 2013
Software complexity metrics are used to quantify a variety of software properties. Complexity measures can be used to predict critical information about testability, reliability and maintainability of the software systems from automatic analysis of the source code. In this paper different software complexity metrics were applied to searching algorithms, our intention is to compare software complexity of linear and binary search algorithms, evaluate, rank competitive object oriented applications (Visual Basic, C#, C++ and Java languages) of these two algorithms using code based complexity metrics such as (line of codes, McCabe cylomatic complexity metrics and Halstead complexity metrics) and measured the sample programs using length (in lines) of the program, line of code (LOC) without comments, LOC with comments, McCabe method, the program difficulty using Halstead method. The result revealed that McCabe method has negligible values of complexity for Visual Basic, C#, C++ and Java languages for linear search and similar measuring values for binary search and also from statistical analysis of ANOVA (Analysis of Variance) the result showed that for both linear and binary search techniques, the four (4) languages do not differ significantly, therefore it is concluded that any of the four (4) programming languages is good to code linear and binary search algorithms.