Software cost estimation models using machine learning and meta heuristic algorithms

Abstract

Meta-heuristic approaches are often better suited to solve real-world problems newlinethan the conventional approach. The solution to every problem cannot always newlinesolve with a precise method, so optimization algorithms are highly effective in newlinethese cases. In order to achieve the desired objective function value, optimization algorithms perform multiple iterations. Therefore, a suitable optimization technique depends on the application s challenges and complexity. Stochastic algorithms must be fast and robust to meet optimization requirements with a high convergence speed, resulting in better cost functions. One real-world problem that many organizations face in the present environment in the field of software engineering is having an efficient model for software cost estimation based on optimization. In this thesis, we design different software cost estimation models based on meta-heuristic and machine learning algorithms newline