Feature Selection and Clustering Algorithms in Data Mining: A Technical Review

Abstract

The partitioning of data into groups of related objects is known as clustering. The data must
obviously lose some fine characteristics when represented by fewer clusters, but
simplification is still achieved. Data is modelled using clusters. Clustering is seen historically
through data modelling, which is based on numerical analysis, statistics, and mathematics.
This paper provides a thorough review of the literature on the most recent developments in
feature selection and data clustering techniques that can be used to learn useful information
about a variety of optimization and clustering issues that can be applied to the processing of
enormous amounts of unstructured and raw data. This document examines and contrasts a
few of the clustering algorithms used by data mining tools currently in use. The important
unsolved issues in this area are then listed, along with the future directions for our research.