• Patchigolla Sahithi
• Dr. B. Sreenivasa Rao
• Dr. Prataprao Prasanna Kumar

Phishing is a severe, complex, and growing menace to online security, easily achievable with deceptive URLs that mislead users regarding sensitive information sharing. This project analyzes the efficacy of machine learning algorithms such as Random Forest, KSTAR, ADABOOST etc in detecting phishing urls. The proposed solution integrates NLP and word vector features to formulate a combined feature set, which has been minimized by PCA to enhance operational speed. The experimental outcome demonstrates Random Forest achieving 99 percent accuracy. For improved classification accuracy, this method adds further method based on an LSTM transformer modeling. This combines a classifier using a specialized dataset of phishing and normal URLs, which outperforms other known techniques by using feature engineering and ensemble classifications. The current work draws attention to the hybridized features and advanced models for automated and robust phishing URL detection.

KSTAR, Phishing, PCA

"URL PHISHING DETECTION WITH PCA-OPTIMIZED HYBRID FEATURES AND ENSEMBLE MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.b140-b148, January-2025, Available :http://www.jetir.org/papers/JETIR2501139.pdf

"URL PHISHING DETECTION WITH PCA-OPTIMIZED HYBRID FEATURES AND ENSEMBLE MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppb140-b148, January-2025, Available at : http://www.jetir.org/papers/JETIR2501139.pdf