Comprehensive analysis of different types of feature selection

Zhanpeng Qi

doi:10.1117/12.2627215

14 February 2022 Comprehensive analysis of different types of feature selection

Zhanpeng Qi

Proceedings Volume 12161, 4th International Conference on Informatics Engineering & Information Science (ICIEIS2021); 121610P (2022) https://doi.org/10.1117/12.2627215
Event: 4th International Conference on Informatics Engineering and Information Science, 2021, Tianjin, China

Abstract

Following the development of research on machine learning in general, feature selection, as a process utilized in machine learning, has also gained some attention. As a common process in machine learning, it would be normal to be enhanced on, just to increase the overall efficiency of the processing. However, to achieve that, one must gain a deep understanding of the topic itself first. In this paper, we will offer a general and comprehensive review and analysis for feature selection. Stating from three basic methods of feature selections: Wrapper, Filter, and embedded, we introduced how those methods work and analyze them as well. Then, we started introducing the steps necessary to perform feature selection: Generation Procedure, Evaluation Function, Stopping Criterion, and Validation Procedure. Finally, we would review and analyze a few recent developments and important concepts related to feature selection: MIC Formulation, Distance Correlation, Model-Based Ranking, Recursive Feature Elimination, and Laplacian score. Those are the methods that represent the recent development of feature selection. This paper is a useful review and reference for people who just started entering the field of feature selection and machine learning in general. It may help one understand some important concept from the ground in feature selection or machine learning in general.

Citation Download Citation

Zhanpeng Qi "Comprehensive analysis of different types of feature selection", Proc. SPIE 12161, 4th International Conference on Informatics Engineering & Information Science (ICIEIS2021), 121610P (14 February 2022); https://doi.org/10.1117/12.2627215

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available