Optimization of Software Project Risk Assessment Using Neuro-Fuzzy Techniques

Abstract

Hazard evaluation assumes a pivotal part in the product venture administration. The discriminating examination of distinctive danger evaluation techniques help specialists and professionals to assess the effect of different venture related dangers. The existing Fuzzy Expert Cost Constructive Model(Fuzzy ExCOM) model is a combination of fuzzy technique and Expert COCOMO. It takes help of mastery and data from prior exercises conveyed for expense and exertion estimation. However, it has limitations that it can't make space for backing from other noteworthy rules related to risks. The proposed work examinations the effect of the ANN technique for software project risk assessment. It serves to create danger standards utilizing Artificial Neural Network techniques to enhance the exactness of danger evaluation process. The combination of various optimization algorithm like Genetic Algorithms and Particle Swarm Optimization are applied collaboratively with Neural network to get best initial starting solution for Neural Network. The results show that this strategy with accessible task information and Neuro-Fuzzy Risk assessment technique provides enhanced outputs than existing Fuzzy Ex-com technique.