PERFORMANCE OPTIMIZATION OF PREDICTIVE MODELS USING PRINCIPAL COMPONENT ANALYSIS AND RANDOM FOREST
DOI:
https://doi.org/10.64751/Abstract
High-dimensional datasets frequently degrade the performance of machine learning models due to their excessive attributes, increased computational complexity, and susceptibility to overfitting. Using all available features to train classifiers makes them run slower and less accurately. In many real-life situations, noisy and unimportant features make the model less accurate and less able to grow. This paper introduces an experimental framework employing Java that integrates ensemble learning and dimensionality reduction techniques to improve classification performance. To start, Principal Component Analysis (PCA) and attribute selection methods are used to reduce the number of dimensions in the dataset while keeping important information. After selecting the optimal reduced dataset, an ensemble classifier known as Random Forest is employed to enhance the accuracy and reliability of the predictions. All tests are done in Java using the WEKA framework. To measure performance, you look at metrics for reducing features, execution time, and accuracy of classification. The results show that combining dimensionality reduction and ensemble learning greatly improves the performance of machine learning as a whole compared to traditional single-model methods.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.






