A Neuro-Fuzzy Model with SEER-SEM for Software Effort Estimation

Software effort estimation is a critical part of software engineering. Although many techniques and algorithmic models have been developed and implemented by practitioners, accurate software effort prediction is still a challenging endeavor. In order to address this issue, a novel soft computing framework was previously developed. Our study utilizes this novel framework to develop an approach combining the neuro-fuzzy technique with the System Evaluation and Estimation of Resource Software Estimation Model (SEER-SEM). Moreover, our study assesses the performance of the proposed model by designing and conducting evaluation with published industrial project data. After analyzing the performance of our model in comparison to the SEERSEM effort estimation model alone, the proposed model demonstrates the ability of improving the estimation accuracy, especially in its ability to reduce the large Mean Relative Error (MRE). Furthermore, the results of this research indicate that the general neuro-fuzzy framework can work with various algorithmic models for improving the performance of software effort estimation.