Flight delays pose a significant challenge for the aviation industry, resulting in substantial financial implications each year. Prior studies have explored machine learning approaches to predict flight delays. However, relying solely on a single airport route may not suffice due to the dynamic nature of the aviation industry. Accurately estimating flight delays is crucial for airlines to enhance customer satisfaction and maximize revenue. This study proposes a novel Deep Learning (DL) model utilizing Support Vector Machine (SVM) for flight delay prediction. DL is a state-of-the-art technique capable of handling complex problems with large datasets and automatically extracting essential features. To address the presence of noisy flight delay data, a stack denoising autoencoder technique is incorporated into the proposed model. The results demonstrate that the proposed model achieves higher accuracy compared to the previous RNN model when forecasting flight delays for imbalanced and balanced datasets. The primary objective is to assess delays and analyse the underlying factors influencing them. The developed system leverages Support Vector Machine, Random Forest, and K-Nearest Neighbours (KNN) algorithms.