Effort analysis and estimation in software projects

This study examines how a classification-based approach is applied to effort estimation in software project management. Aimed at achieving faster, more consistent, and repeatable effort estimation in software projects, which are mostly reliant on manual and expert judgment, this study proposes a system where software companies can save time and costs while obtaining consistent and reliable estimates. Within the scope of the research, two different classification models were developed for software effort estimation using the logistic regression algorithm. The goodness-of-fit statistics of the constructed model and the classification success of the model were analyzed, revealing high explanatory and accuracy rates. Additionally, the effects of data preprocessing on the classification model were investigated, and comparisons were made with studies in the literature using the same dataset.