Software development effort estimation modeling using a combination of fuzzy-neural network and differential evolution algorithm

Software cost estimation has always been a serious challenge lying ahead of software teams that should be seriously considered in the early stages project. Lack sufficient information on final requirements, as well existence inaccurate and vague are among main reasons for unreliable estimations this area. Though several effort models have proposed over recent decade, an increase their accuracy controversial issue, researchers' efforts area still ongoing. This study presents new model based hybrid adaptive network-based fuzzy inference system (ANFIS) differential evolution (DE) algorithm. tries to obtain more accurate development is capable presenting better estimate within wide range projects compared previous works. The method outperformed other optimization algorithms adopted from genetic algorithm, evolutionary algorithms, meta-heuristic neuro-fuzzy could improve using MMRE PRED (0.25) criteria up 7%.