Exploring the Regression Path of Deep Learning Algorithms for Big Data and High-Dimensional Data

Abstract

Deep learning algorithms have become crucial for handling and extracting insights from big data and high-dimensional data. This paper explores the regression capabilities of Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory Networks (LSTMs) in predicting patient outcomes in a hospital setting. By leveraging these advanced algorithms, the study aims to automate feature extraction, thereby reducing the need for manual feature engineering. The models were evaluated using standard regression metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R-squared (R²) score. The results indicate that LSTMs outperform CNNs and RNNs across all metrics, highlighting their superior ability to capture complex temporal patterns in high-dimensional medical data. This study underscores the potential of deep learning algorithms in enhancing predictive accuracy and operational efficiency in healthcare.