Software Effort Estimation Using Modified Fuzzy C Means Clustering and Hybrid ABC-MCS Optimization in Neural Network

In a software development process, effective cost estimation is the most challenging activity. Software effort estimation is a crucial part of cost estimation. Management cautiously considers the efforts and benefits of software before committing the required resources to that project or order for a contract. Unfortunately, it is difficult to measure such preliminary estimation, as it has only little information about the project at an early stage. In this paper, a new approach is proposed; this is based on reasoning by the soft computing approach to calculate the effort estimation of the software. In this approach, rules are generated based on the input dataset. These rules are then clustered for better estimation. In our proposed method, we use modified fuzzy C means for clustering the dataset. Once the clustering is done, various rules are obtained and these rules are given as the input to the neural network. Here, we modify the neural network by incorporating optimization algorithms. The optimization algorithms employed here are the artificial bee colony (ABC), modified cuckoo search (MCS), and hybrid ABC-MCS algorithms. Hence, we obtain three optimized sets of rules that are used for the effort estimation process. The performance of our proposed method is investigated using parameters such as the mean absolute relative error and mean magnitude of relative error.