Big Data-Driven Phishing Detection in Smart Devices Using Chi-Square and Optimized Gradient Boosting

doi:10.1007/978-981-96-6294-4_16

Big Data-Driven Phishing Detection in Smart Devices Using Chi-Square and Optimized Gradient Boosting

2026-01-01

0

PRE

AI

DOI：10.1007/978-981-96-6294-4_16

Original

Original request for help

Save

Abstract

En 中文

Phishing attacks are a major cybersecurity threat, especially in smart devices, where attackers exploit vulnerabilities to steal sensitive information. As the complexity of phishing techniques grows, the need for robust detection methods becomes critical. This paper presents a Big Data based model to identify phishing in smart devices. Following PySpark for data preprocessing and Chi-Square feature selection, the suggested model optimizes a Gradient Boosting model using the Probabilistic Bees Algorithm (BeesA). With high accuracy, recall, and Fl score, the model was assessed on a dataset including more than 11,000 webpages. Comparative study with conventional classifiers like Random Forest, SVM, and Naive Bayes shows the better performance of the suggested model. The results show how well integrating Big Data methods with sophisticated optimization algorithms improves phishing detection in smart device scenarios.

Keywords:

Phishing Detection

Big Data

Smart Devices

Gradient Boosting

Probabilistic Bees Algorithm

Big Data-Driven Phishing Detection in Smart Devices Using Chi-Square and Optimized Gradient Boosting

Abstract

Journal

Organization