Melodic Precision: Unveiling the Impact of Recursive Feature Elimination on Music Genre Classification

Music genre classification has experienced significant advancements due to the integration of machine learning techniques. This study delves into the impact of the Recursive Feature Elimination (RFE) technique on the performance of music genre classification models. Utilizing the GTZAN dataset, which includes various audio features, we implement machine learning models such as Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, and Support Vector Machines (SVM). By employing the RFE technique, we aim to enhance model accuracy and interpretability. Analyzing classification metrics such as accuracy, precision, recall, F1 score, and confusion matrices, we provide a nuanced understanding of how feature selection and elimination using RFE refine the accuracy and interpretability of music genre classification models. Our findings reveal the effectiveness of RFE in improving the discriminative power of music genre classifiers, highlighting the importance of individual feature contributions. Additionally, our study compares the impact of RFE with other feature selection and elimination techniques such as variance threshold, Chisquare test, and information gain, offering valuable insights into optimizing music genre classification models for practical applications. These results underscore the potential of RFE to advance the field of music genre classification by improving model performance and interpretability, making it a valuable tool for researchers and practitioners in the domain.