An Exploratory Study of Cognitive Based Complexity Measures of Online Algorithms

Measuring the complexity of software has been an insoluble problem in software engineering. Complexity measures can be used to predict critical information about testability of software system from automatic analysis of the source code. In this paper, Improved Cognitive Complexity Metric (ICCM) is applied on C programming language. Since C is a procedural language, the cognitive complexity metric is capable to evaluate any procedural language. This paper presents a cognitive complexity metric named ICCM. First, the metric is analytically evaluated using Weyuker’s properties for analyzing its nature. Secondly, perform a comparative study of the metric with the existing metric such as NCCOP, CFS, CICM and CPCM, and the result shows that ICCM does better than other metrics by giving more information contained in the software and reflecting the understandability of a source code. Also, an attempts has also been made to present the relationship among ICCM, NCCOP, CICM, CFS and CPCM using pearson correlation coefficient method.