| This paper studies the Hierarchical Clustering-Ensemble Regressor (HCER) framework, a novel and advanced approach in predictive analysis that combines hierarchical clustering with ensemble methods. The HCER framework is designed to leverage the intricate structures inherent in large-scale data, often overlooked by standard regression models. By identifying and utilizing sub-populations within datasets, this approach enables the discovery of unique predictive dynamics and enhances the accuracy of predictions. Our study extends the application of the HCER framework beyond its initial deployment in the ATHLOS dataset, exploring its versatility and effectiveness across various data domains. We highlight the capability of the framework to capture localized patterns in data, which might be lost in global models, thereby enriching the understanding of the data's underlying structure. This is crucial in fields where deep data comprehension is as essential as prediction accuracy. Additionally, we investigate the computational efficiency of the HCER framework, considering the growing need for scalable and efficient methods in the era of Big Data. Our findings indicate that the HCER framework contributes to the machine learning domain by enhancing prediction accuracy and offering a scalable solution suitable for large datasets. This comprehensive study positions the HCER framework as a potent predictive modeling tool capable of handling the diversity and complexity of contemporary data landscapes. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.