<![CDATA[This paper examines a more accurate and broader classification model and has significant implications in these fields. Combining multiple models or using hybrid models has become common practice to overcome the shortcomings of a single model and can be a more effective way to improve its predictive performance, especially when the models are in very different combinations. In this paper, a new hybridization of artificial neural networks (ANN) is proposed using multiple linear regression models to produce more accurate models than traditional artificial neural networks for solving classification problems. Empirical results show that the proposed hybrid model shows to effectively improve classification accuracy compared to traditional artificial neural networks and also several other classification models such as linear discriminant analysis, quadratic discriminant analysis, and vector machine using benchmarks and real-world application datasets. These datasets vary in number of classes and data sources. Therefore, it can be applied as a suitable alternative approach to solve classification problems, especially when higher forecasting accuracy is required.]]>
Copyrights © 2024
