Introduction

We find ourselves immersed in an era defined by the exponential growth of data, encompassing images, videos, and documents. As the volume of data escalates, the extraction of numerous features becomes necessary, leading to the challenge known as the curse of dimensionality. Within this high-dimensional data lie redundant information and concealed correlations, surpassing the capabilities of traditional manual processing. In the domains of pattern recognition and data mining, dimension reduction and data clustering emerge as pivotal learning techniques. Dimension reduction seeks to project data from high-dimensional spaces into lower-dimensional spaces, yielding a more concise and compact representation. By reducing the complexity of data processing and facilitating the discovery of data structure information, dimension reduction enables enhanced visualization and accelerates data analysis.