Human Opinion Dynamics for Software Cost Estimation

Human opinion dynamics is a novel approach to solve complex optimization problems. In this paper we propose and implement human opinion dynamics for tuning the parameters of the COCOMO model for software cost estimation. The input is coding size or lines of code and the output is effort in person-months. Mean absolute relative error and prediction are the two objectives considered for fine-tuning of parameters. The dataset considered is COCOMO. The current paper demonstrates that use of human opinion dynamics illustrates promising results. It has been observed that when compared with standard COCOMO it gives better results.