Parameter Identification of a Multilayer Perceptron Neural Network using an Optimized Salp Swarm Algorithm

Effort estimation in software development (SEE) is a crucial concern within the engineering domain, as it directly impacts cost estimation, scheduling, staffing, planning, and resource allocation accuracy. In this scientific article, authors aim to tackle issue by integrating machine learning (ML) techniques with metaheuristic algorithms order raise prediction For purpose, they employ multilayer perceptron neural network (MLP) perform for SEE. Unfortunately, MLP has numerous drawbacks well, including weight dependency, rapid convergence, accuracy limits. To address these issues, SSA Algorithm employed optimize weights biases. Simultaneously, algorithm shortcomings some aspects of search mechanisms such convergence being susceptible local optimal trap. As result, genetic (GA) utilized through fine-tuning its parameters. The main objective develop robust reliable model that can handle wide range SEE problems. developed are tested on twelve benchmark datasets evaluate their performance. Furthermore, comparative analysis state-of-the-art methods conducted further validate effectiveness techniques. findings demonstrate surpass all other problems, affirming superiority.