Prediction of Building Permit Approval using CNN

Efficient prediction of building permit approvals is essential for urban planning and construction management. This research investigates the comparative performance of Machine Learning (ML) and Deep Learning (DL) models in predicting permit approvals based on a large-scale dataset. Different models, such as Logistic Regression, Support Vector Machine (SVM), Decision Tree, Random Forest, XGBoost, and Convolutional Neural Network (CNN), were applied and compared. The findings show that CNN performed best among all the models, with a maximum accuracy of 90.5%, showcasing its capacity to learn intricate spatial and textual features from permit datasets. Performance measures in terms of accuracy, precision, recall, and F1-score were examined, as well as confusion matrices and trends in training accuracy. The results indicate that deep learning-based models can dramatically improve prediction accuracy in permit approval systems, minimizing manual effort and enhancing decision-making processes.