Mathematical Methods in Interdisciplinary Sciences

9Fuzzy Rough Set Theory‐Based Feature Selection: A Review

Tanmoy Som¹, Shivam Shreevastava², Anoop Kumar Tiwari³, and Shivani Singh⁴

¹Department of Mathematical Sciences, Indian Institute of Technology (Banaras Hindu University), Varanasi, Uttar Pradesh, 221005, India

²Department of Mathematics, SBAS, Galgotias University, Gautam Buddha Nagar, Uttar Pradesh, 201310, India

³Department of Computer Science and Engineering, Dr. K. N. Modi University, Tonk, Rajasthan, 304021, India

⁴DST‐Centre for Interdisciplinary Mathematical Sciences, Institute of Science, Banaras Hindu University, Varanasi, Uttar Pradesh, 221005, India

9.1 Introduction

Dimensionality reduction [1–3] has been an intriguing field of research in machine learning, signal processing, medical image processing, bioinformatics, etc. The generation of large volume of real‐valued datasets from various domains increases the requirement of dimensionality reduction in order to produce the most informative features. Dimensionality reduction likewise improves the performances of fast storage systems as well as prediction algorithms. Feature selection is one of the dimensionality reduction procedures, which preserves the striking qualities of the database systems. It shrinks exceptionally highly correlated features, which may result in lowering the overall accuracy. Attribute selection strategies center around progressively significant and non‐redundant features. It has been implemented in numerous areas such as text categorization ...