Addressing Multiple Linear Regression Challenges with Artificial Neural Networks: A Comparative Study
Author : R Pushpalatha, G Ranipadmini and V Gomathi
Abstract :
Linear regression is widely used for predictive analysis, but solving multiple linear regression models separately can be time-consuming and computationally inefficient. This paper explores the application of Artificial Neural Networks (ANNs) to handle multiple linear regression tasks simultaneously. By leveraging the parallel processing capabilities of ANNs, we transform multiple regression problems into a single, unified framework. Experimental results demonstrate that ANNs not only match traditional linear regression in accuracy but also handle complex, high-dimensional data more efficiently. This approach reduces computational complexity, improves scalability, and offers a streamlined solution for multi-task regression, benefiting fields like finance, healthcare, and engineering.
Keywords :
Artificial neural network, Convolutional artificial neural network, feed forward artificial neural network and multiple linear regression
