Leveraging PCA to reduce Complexity and Converge Accuracy When Predicting Illnes

High-dimensional datasets often pose challenges for machine learning models, including increased computational complexity, noise, and the curse of dimensionality. Principal component analysis (PCA) is a well-established technique for dimensionality reduction that addresses these issues by projecting the data into a lower-dimensional space while retaining most of the variance. In this project, we set out to build a baseline neural network model in order to establish a performance benchmark on the high-dimensional dataset. Subsequently, PCA is applied to reduce the dimensionality of the input features, mitigating noise and improving efficiency during inference. The resulting low-dimensional representation is used to train a second neural network, demonstrating the trade-offs between dimensionality reduction, model complexity, and predictive performance. This approach highlights the synergy between PCA and neural networks for efficient and robust modeling in highdimensional settings