Welcome to the InfoSci Platform
Reference Hub2
A Study of Feature Selection and Dimensionality Reduction Methods for Classification-Based Phishing Detection System

A Study of Feature Selection and Dimensionality Reduction Methods for Classification-Based Phishing Detection System

Amit Singh, Abhishek Tiwari
Copyright: © 2021 |Volume: 11 |Issue: 1 |Pages: 35
ISSN: 2155-6377|EISSN: 2155-6385|EISBN13: 9781799861980|DOI: 10.4018/IJIRR.2021010101
Cite Article Cite Article

MLA

Singh, Amit, and Abhishek Tiwari. "A Study of Feature Selection and Dimensionality Reduction Methods for Classification-Based Phishing Detection System." IJIRR vol.11, no.1 2021: pp.1-35. http://doi.org/10.4018/IJIRR.2021010101

APA

Singh, A. & Tiwari, A. (2021). A Study of Feature Selection and Dimensionality Reduction Methods for Classification-Based Phishing Detection System. International Journal of Information Retrieval Research (IJIRR), 11(1), 1-35. http://doi.org/10.4018/IJIRR.2021010101

Chicago

Singh, Amit, and Abhishek Tiwari. "A Study of Feature Selection and Dimensionality Reduction Methods for Classification-Based Phishing Detection System," International Journal of Information Retrieval Research (IJIRR) 11, no.1: 1-35. http://doi.org/10.4018/IJIRR.2021010101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Phishing was introduced in 1996, and now phishing is the biggest cybercrime challenge. Phishing is an abstract way to deceive users over the internet. Purpose of phishers is to extract the sensitive information of the user. Researchers have been working on solutions of phishing problem, but the parallel evolution of cybercrime techniques have made it a tough nut to crack. Recently, machine learning-based solutions are widely adopted to tackle the menace of phishing. This survey paper studies various feature selection method and dimensionality reduction methods and sees how they perform with machine learning-based classifier. The selection of features is vital for developing a good performance machine learning model. This work is comparing three broad categories of feature selection methods, namely filter, wrapper, and embedded feature selection methods, to reduce the dimensionality of data. The effectiveness of these methods has been assessed on several machine learning classifiers using k-fold cross-validation score, accuracy, precision, recall, and time.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.