Cyber Threat Detection Using an Ensemble Model Approach for Phishing Website Identification

The development of digital technology has had a significant impact on various aspects of life, including an increase in cybersecurity threats, especially phishing attacks. Phishing is a method of cyber fraud that manipulates victims to provide sensitive information by posing as a trusted entity. This research aims to develop and evaluate the effectiveness of several machine learning algorithms in detecting phishing websites. The methods used in this research include the application of Random Forest, Extra Trees, Multiple Layer Perceptron, Ada Boost, and Decision Tree algorithms on website datasets containing the characteristics of phishing and non-phishing sites. Performance evaluation is performed by measuring the accuracy, precision, recall, and F1 value of each algorithm. In addition, a voting technique is applied to combine the results of the best-performing algorithms with the aim of improving the overall detection accuracy. The results showed that the voting technique was able to provide superior results compared to the use of a single algorithm, with significant improvements in accuracy and recall values. These findings reinforce the importance of ensemble approaches in machine learning to improve phishing detection capabilities, which in turn contributes to improved cybersecurity.