A Survey of Machine Learning Approach to Software Cost Estimation

Software companies are growing fast every day due to the high demand for software, while software development cost increasing at the same time. In order to overcome this concern, we require an efficient technique for more accurately estimating software costs to manage and control the costs and further make the software more reliable and competitive. Software development by default is a challenging process that may face deep and essential problems especially when we trying to create accurate and reliable software cost estimates. These issues are strengthening due to the high level of difficulty, and complexity of the software process. The goal of this study is to address the difficulties of estimating the software development cost using conventional approaches. Further identifying the necessary steps for computable entities which affect the software cost and presenting the research works that utilize them with machine learning approaches to build a reliable estimation method. The various proposed software cost estimation methods with distinct designs assessed and collected outcomes reported.