• JOSEPH R

Music spans a wide spectrum of styles, from classical and jazz to rock, pop, and folk. At the heart of understanding harmony lies chord recognition, a crucial task for both music analysis and performance. While modern audio processing techniques such as chromagrams, MFCCs, and pitch class profiles have made chord recognition feasible, they also produce large, high-dimensional feature spaces. These high-dimensional representations can introduce redundancy, slow down computation, and reduce classification accuracy due to overfitting. To address these challenges, this study explores the role of dimensionality reduction in chord recognition. The proposed system processes audio recordings into feature vectors and applies three reduction techniques: Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Uniform Manifold Approximation and Projection (UMAP). These transformed features are then classified using machine learning models such as Random Forest, Gradient Boosting, Logistic Regression, and Support Vector Machines. Experiments conducted on annotated chord datasets show that not all dimensionality reduction techniques contribute equally. While PCA and UMAP significantly degraded recognition accuracy, LDA improved classification results, achieving a peak accuracy of 94.21% with Random Forest. This demonstrates that supervised dimensionality reduction, which emphasizes class separability, is particularly effective for chord recognition tasks.

Music Information Retrieval, Chord Recognition, Dimensionality Reduction, PCA, LDA, UMAP, Machine Learning

"Dimensionality Reduction for Chord Recognition Features in Music Information Retrieval", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 9, page no.d488-d492, September-2025, Available :http://www.jetir.org/papers/JETIR2509365.pdf

"Dimensionality Reduction for Chord Recognition Features in Music Information Retrieval", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 9, page no. ppd488-d492, September-2025, Available at : http://www.jetir.org/papers/JETIR2509365.pdf