Phishing Detection System Through Hybrid Machine Learning Based on URL

Currently, numerous types of cybercrime are organized through the internet. Hence, this study mainly focuses on phishing attacks. Although was first used in 1996, it has become most severe and dangerous Phishing utilizes email distortion as its underlying mechanism for tricky correspondences, followed by mock sites, to obtain required data from people question. Different studies have presented their work precaution, identification, knowledge attacks; however, there is currently no complete proper solution frustrating them. Therefore, machine learning plays a vital role defending against cybercrimes involving The proposed based URL-based dataset extracted famous repository, which consists legitimate URL attributes collected 11000+ website datasets vector form. After preprocessing, many algorithms been applied designed prevent URLs provide protection user. This uses models such decision tree (DT), linear regression (LR), random forest (RF), naive Bayes (NB), gradient boosting classifier (GBM), K-neighbors (KNN), support (SVC), hybrid LSD model, combination logistic regression, machine, (LR+SVC+DT) with soft hard voting, defend attacks high accuracy efficiency. canopy feature selection technique cross fold valoidation Grid Search Hyperparameter Optimization techniques model. Furthermore, evaluate approach, different evaluation parameters were adopted, precision, accuracy, recall, F1-score, specificity, illustrate effects efficiency models. results comparative analyses demonstrate that approach outperforms other achieves best results.