Hybrid Approach for Phishing Website Detection Using Classification Algorithms

ParadigmPlus

article

Mukta MithraRaj¹  J. Angel Arul Jothi¹ 

[1] Birla Institute of Technology and Science Pilani

关键词: URL Features;    Data Mining;    Machine Learning;    Hybrid Classification Algorithms;    Phishing Website Detection;   

DOI  :  10.55969/paradigmplus.v3n3a2

学科分类：环境工程

来源: ITI Research Group

【 摘 要 】

The internet has significantly altered how we work and interact with one another. Statisticsshow 63.1% of the present world population are internet users. This clearly indicates how heavilyman is dependent on digital media. Digital media users are on the rise and so is the incidence ofcybercrimes. People who lack experience and knowledge are more vulnerable and susceptible tophishing scams. The victims experience severe consequences as their personal credentials are atstake. Phishers use publicly available sources to acquire details about the victim’s professional andpersonal history. Countermeasures must be implemented with the highest priority. Detection ofmalicious websites can significantly reduce the risk of phishing attempts. In this research, a highlyaccurate website phishing detection method based on URL features is proposed. We investigatedeight existing machine learning classification techniques for this, including extreme gradient boosting (XGBoost), random forest (RF), adaptive boosting (AdaBoost), decision trees (DT), K-nearestneighbors (KNN), support vector machines (SVM), logistic regression and Naive Bayes (NB) todetect malicious websites. The results show that XGboost had the best accuracy with a score of96.71%, followed by random forest and AdaBoost.We further experimented with various combinations of the top three classifiers and observed that XGboost-Random Forest hybrid algorithmsproduced the best results. The hybrid model classified the websites as legitimate or phishing withan accuracy of 97.07%.

【 授权许可】

CC BY