GUARDING THE GATE: Innovative Solutions for Third-Party App Vulnerabilities

Phishing poses a critical concern owing to the hike in the number of phishing instances and sophisticated tactics used by hackers to attack in an unlawful manner. For anyone to identify the portals indulging in the phishing activities, a few phishing prohibition approaches need to be adopted. These approaches often take advantage of 3rd party services and source code-based features. Subsequently, we propose a comprehensive technique for assessing the security of 3rd party apps by adopting both the malware and phishing website recognitions. The main objective of this work is to determine the safety status of 3rd party app, thereby providing users with insights for acting towards the security conduct of any considered systems. To perform recognition of the malware, we are going to assess the performance of methodologies like Random Forest, Decision Trees, AdaBoost, and XGBoost with the help of a few common performance metrics like accuracy, recall, precision, F1 Score, and ROC-AUC curves. To perform recognition of the website attempting phishing activities, we are employing methodologies like Random Forest, Gradient Boosting, XGBoost, AdaBoost, and Support Vector Classification (SVC). Similar to malware recognition, we are measuring the performance using the performance metrics like precision, accuracy, recall, F1 Score, and ROC-AUC curves. Finally, we validated the obtained results and discussed the insights out of the work.