Phishing Attack Detection using a Search Engine and Heuristics-based Technique
Brij B. Gupta, Ankit Kumar Jain
Abstract
The language used in the textual content of the webpage is the barrier in most of the existing anti-phishing methods. Most of the existing anti-phishing methods can identify the fake webpages written in the English language only. Therefore, we present a search engine-based method in this article, which identifies phishing webpages accurately regardless of the textual language used within the webpage. The proposed search engine-based method uses a lightweight, consistent and language independent search query to detect the legality of the suspicious URL. We have also integrated five heuristics with the search engine-based mechanism to improve the detection accuracy, as some newly created legitimate sites may not appear in the search engine. The proposed method can also correctly classify the newly created legitimate sites that are not classified by available search engine-based methods. Evaluation results show that our method outperforms the available search-based techniques and achieves 98.15% TPR of and only 0.05% FPR.