Provided by: International Journal of Computer Applications
To develop a project successfully, it is important for any organization that the project should be completed within budget, on time and the project should have requisite quality. This paper presents an adaptive Neuro-Fuzzy approach for software development time estimation. This proposed technique is aimed at building and evaluating a Neuro-Fuzzy model using three Membership Functions (MFs) for software project development time. The forty one modules were used as a data set. The authors' proposed approach for Neuro-Fuzzy using 3 membership functions i.e. Gaussian MF (GMF), Triangular MF (Tri MF) and Trapezoidal MF (Trap MF) is compared with neural network models and the results show that values of various relative error parameters for Neuro-Fuzzy is lower than the values of parameters applying neural network.