Phishing Detection

Phishing attacks continue to be a major security threat for individuals and organizations alike. It causes billions of dollars in losses annually. Machine learning(ML) has shown great promise in detecting such attacks by identifying patterns and anomalies in large datasets. The tradeoff between feature selection and model selection is a tedious task in ML for phishing detection. Low number of features are not enough for the generalizability of traditional machine learning algorithms i.e.For Logistic Regression(LR), Support Vector Machine(SVM), Random Forest(RF), XG boost and Naive Bayes(NB). And it’s tough for deep learning(DL) algorithms to learn features from ambiguities behaviour between phishing and non-phishing websites. This paper presents a comprehensive survey of various ML and ML paradigms that have been employed for the detection of phishing websites. The survey also discusses various datasets, features, number of parameters in algorithms, training time-space complexity in phishing detection and compares the accuracies of different ML techniques. The results of this survey provide valuable insight into the current state of the art in phishing detection and can serve as a useful resource for researchers and practitioners working in this field.

Run Project

The Github repository can also be found here:

cd phishing-detection-transformer

cd src

pip3 install -r requirements.txt

streamlit run app.py