Development of Proposed Model Using Random Forest with Optimization Technique for Classification of Phishing Website

• Published in: SN computer science Vol. 5; no. 8; p. 1059
• Main Authors: Pathak, Prakash, Shrivas, Akhilesh Kumar
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.12.2024; Springer Nature B.V
• Subjects: Accuracy; Classification; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Computing time; Cybercrime; Cybersecurity; Data Structures and Information Theory; Datasets; Decision trees; Ensemble learning; Feature selection; Genetic algorithms; Information Systems and Communication Service; Literature reviews; Machine learning; Optimization techniques; Original Research; Particle swarm optimization; Pattern Recognition and Graphics; Phishing; Research Advancements in Intelligent Computing; Software Engineering/Programming and Operating Systems; Support vector machines; Trends; Vision; Websites; Proposed Model; Phishing Website; Feature Selection Technique (FST); Classification; Particle Swarm Optimization (PSO); Random Forest
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-024-03388-x

A phishing website is a fraudulent online platform intentionally created to mimic trustworthy websites to steal private and sensitive data from unwary users. The word “phishing” comes from the word “fishing,” whereby online thieves utilize fake websites as bait to trick people into giving up personal information like passwords, usernames, and bank account information. Phishing websites use social engineering techniques to generate a false sense of urgency or anxiety. They are characterized by a deceptive design that mimics genuine websites and URL manipulation through subtle misspellings or domain variations. Phishing attacks frequently start with false emails, messages, websites, or advertisements that contain links that take visitors to these hazardous websites. This research paper focuses on phishing website classification using machine learning based classification techniques with Particle Swarm Optimization (PSO) feature selection technique. We have used different classification techniques like K-Nearest Neighbours (K-NN), Gaussian Naïve Bayes (GNB), Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Adaptive Boosting (AdaBoost), and ensemble classifiers for classification of phishing websites. We have used PSO feature selection technique to reduce the features from phishing website dataset. The main aim of PSO feature selection technique is to computationally increase the performance of the model and improve the classification accuracy. We have also compared the performance measures of classifiers or models with and without feature selection technique where our proposed RF-PSO model achieves a better performance in terms of accuracy as 97.84%, Recall as 99.00%, and F1-score as 98.69% with 14 features and less computational time.