The impact of the additional features on the performance of regression analysis: a case study on regression analysis of music signal
This addresses the dilemma of algorithm selection in complex tasks for practitioners in data science and music signal processing, though it appears incremental in its approach.
The paper investigates whether simple statistical machine learning algorithms can outperform deep learning techniques by focusing on data preprocessing, specifically using trigonometric, logarithmic, and exponential functions, and applies this to regression analysis on music signals to demonstrate performance improvements.
Machine learning techniques nowadays play a vital role in many burning issues of real-world problems when it involves data. In addition, when the task is complex, people are in dilemma in choosing deep learning techniques or going without them. This paper is about whether we should always rely on deep learning techniques or it is really possible to overcome the performance of deep learning algorithms by simple statistical machine learning algorithms by understanding the application and processing the data so that it can help in increasing the performance of the algorithm by a notable amount. The paper mentions the importance of data preprocessing than that of the selection of the algorithm. It discusses the functions involving trigonometric, logarithmic, and exponential terms and also talks about functions that are purely trigonometric. Finally, we discuss regression analysis on music signals to justify our claim.