Design and Development of enhanced N-dimensional data reduction system using Discrete Cosine Transform (DCT) and Polynomial Regression Analysis (PRA) for modelling

In the era of big data and high-dimensional datasets, efficient data reduction techniques are crucial for improving computational efficiency and ensuring model interpretability. This study presents the design and development of an enhanced N-dimensional data reduction system that integrates Discrete Cosine Transform (DCT) and Polynomial Regression Analysis (PRA) to address challenges in data redundancy and noise. DCT is employed to compress and transform complex high-dimensional data into a compact representation, while PRA is applied to retain underlying trends and relationships among variables. The proposed hybrid framework not only reduces dimensionality but also maintains high prediction accuracy and fidelity in downstream modelling tasks. The system's performance is evaluated across synthetic and real-world datasets, demonstrating significant improvements in computation time, storage efficiency, and modelling accuracy when compared to traditional techniques like PCA and linear regression. This research provides a scalable and adaptable solution for applications in environmental modelling, image processing, and sensor-based data systems.