Effort estimation is essential for successful software project planning,
budgeting, and risk identification, ensuring projects are completed within budget, on time, and to the required quality. However, the methods used to estimate effort are often inaccurate and outdated and only consider technical factors, ignoring non- technical ones like project management or stakeholder engagement. Nevertheless, traditional expert estimation remains an important technique for leveraging human expertise in software estimation but solely relying on this technique causes biased and subjective predictions. Some machine learning algorithms have drawn the attention of researchers because of their data-driven approach and have changed the direction of software effort estimation toward computational intelligence. Nonetheless, mostly due to ambiguous outcomes and vague model-building approaches, there are barely any deployments in practice. This paper presents an improved machine-learning model that can predict software effort estimation more accurately. Furthermore, this study also aims to incorporate domain knowledge and experiences into the model. To achieve this, six homogeneous classifier ensembles will be constructed using six distinct base learner classifiers on the USP05-FT dataset. Subsequently, through majority voting, the predictions of all homogeneous classifier ensembles will be combined to get a more accurate and reliable prediction with increased robustness against errors and uncertainties. Once the ensemble model is constructed, then the collected expert estimations on the proposed USP05-ft datasets will be used as an additional feature in the dataset in the form of numerical values that provides additional insight and knowledge that may not be captured. The performance of the proposed approach will be evaluated by employing parameters such as recall, f-measure, precision, and accuracy. It is expected that the proposed model for software effort prediction will result in improved accuracy, ultimately leading to more efficient and effective software development projects.