Software cost estimation with fuzzy models
ACM SIGAPP Applied Computing Review
Software Engineering Economics
Software Engineering Economics
Improving analogy software effort estimation using fuzzy feature subset selection algorithm
Proceedings of the 4th international workshop on Predictor models in software engineering
Hi-index | 0.00 |
In this research, it is investigated the precision of size and cost drivers in the estimation of effort using Constructive Cost Model (COCOMO). It is imperative to stress that uncertainty at the input level of the COCOMO yields uncertainty at the output, which leads to gross estimation error in the effort estimation. Instead of using a single number to represent the size, it can be characterized as a fuzzy value. Cost drivers also expressed through an unclear category which needs subjective assessment. Fuzzy logic has been applied to the COCOMO using the symmetrical triangles and trapezoidal membership functions to represent the cost drivers and size. Using trapezoidal membership function for the size and cost drivers, a few attributes are assigned the maximum degree of compatibility when they should be assigned lower degrees. To overcome the above limitations, in this work, it is concentrated to use Gaussian membership function for the COCOMO parameters. In addition, this paper proposes to incorporate both size and cost drivers together, with a fuzzy set using Gaussian membership function. The present work is based on COCOMO dataset and the experimental part of the study illustrates the approach and compares it with the standard version of the COCOMO. It has been found that the proposed method is performing better than ordinal COCOMO and the achieved results were closer to the actual effort.