Article Preview
Top2. Literature Survey
In this section various literature works through different peoples were summarized.
Here,(Pujara & Chaudhari, 2018) the methodologies-Heuristic based method, Blacklist method, Visual similarity were used. If the URLs that were caught inside the databases were found, it was marked as a phishing URL and showed warnings, if not it was seen as legitimate site. The heuristic method, identifies phishing potentials by applying features collected through phishing websites for detecting fraud attacking. Visual similarity extracts images which are found in legitimate sites. The approaches performed efficiently in large datasets as was seen in (Pujara & Chaudhari, 2018). Were as, (Revoredo da Silva et al., 2020) shows that new heuristic approaches can be developed through the evaluation of the results which can improve the durability, stability and performance of the existing ones described in (Revoredo da Silva et al., 2020).
The paper (Jain & Gupta, 2017) shows several procedures for phishing detections, with limitations, while detecting embedded objects, identifying fraud websites, countermeasures towards newly built phishing sites. Different features for a webpage to detect fraudulence were utilized like text similarity, font size, font color. The similarity approach of text basing could not detect phishing sites if it interchanged texts with images instead. The other approach for such drawbacks was Image processing-based algorithms. But, as it becomes way complex, it was unable to be applied in (Jain & Gupta, 2017)