An improved technique for software cost estimations in agile software development using soft computing
Sudhir Kumar, Maheshwari Prasad Singh · IET conference proceedings. · 2023
Several software businesses are being encouraged to work because of their high failure rates to manage and estimate agile projects and create successful software projects. The process of accurately estimating a project's overall effort and expense is crucial. Throughout the past few decades, software initiatives have helped to increase the variety of development methodologies. Therefore, it is a big problem to determine the exact amount of effort and cost incurred for various program undertakings that are based on unique improvement models and have fresh and innovative software development processes. Software businesses have used several distinct development approaches, including 1The COCOMO (Constructive Cost Model), which enables more precise cost estimation of software projects, was introduced in this study. It is feasible to forecast the financial success or failure of a new endeavor using historical company data. The Nave Bayes method is crucial and offers good accuracy in machine learning that uses historical data to predict the future. To test the suggested system's behavior in this case, we used the SEERA dataset. The results show that the accuracy of our proposed method in predicting profit and loss is 89.59% and 26.8 percent, respectively. The accuracy of the overall effort computation is also higher, at 97.06 percent, when compared to the SVM's accuracy of 94.45 percent.