Detecting Phishing Web Pages with Visual Similarity Assessment Based on Earth Mover's Distance (EMD)

By Anthony Y. Fu, Wenyin Liu, Xiaotie Deng. 2006

355 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Battista Biggio
2. Fabio Roli
3. Igino Corona
4. Bin Liang
5. Miaoqiang Su
6. Wenchang Shi
7. Gang Yang
8. Wei You

In Information Systems

1. Battista Biggio
2. Fabio Roli
3. Igino Corona
4. Bin Liang
5. Miaoqiang Su
6. Wenchang Shi
7. Wei You
8. Gang Yang

Citations in the corpus

• DeltaPhish: Detecting Phishing Webpages in Compromised Websites (2017, background): “…snapshots or embedded images, using a wide range of image analysis techniques, mostly based on computing low-level visual features, including color histograms, two-dimensional Haar wavelets, and other well-known image descriptors normally exploited in the field of computer vision [30, 31, 12, 13].”
• An Anti-Phishing System Employing Diffused Information (2014, methodology): “The Normalized Compression Distance By de.nition, the Kolmogorov Complexity of a string is the length of the shortest pro­gram that outputs the string; it is known to be incomputable except in trivial cases.”
• Emerging phishing trends and effectiveness of the anti-phishing landing page (2014, background): “8 To see if phishers incorporated similar technique, we studied if there exists any correlation between the location from where user clicked the phishing URL and the language in which the landing page was displayed.”
• Phishing Detection: A Literature Survey (2013, methodology): “The visual similarity technique functions as proposed previously in [43], which takes snapshots of websites as rendered by a web browser (Internet Explorer in their case), transforms them to 100 x 100 pixels images, generates a features vector, and then returns a normalized number that reflects the ”
• Feature selection for text classification: A review (2018, methodology): “…measures for text processing, for example, Euclidean distance, Jaccard coefficient [54], Pearson correlation [40], Cosine Similarity [6], Hamming distance [62], Dice coefficient [85], IT-Sim [3, 56], SMTP [58], Earth mover’s distance [29, 96], Kullback-Leibler divergence [50], and BM25 [74].”
• PhishZoo: Detecting Phishing Websites by Looking at Them (2011, methodology): “used Earth Mover’s Distance (EMD) to compare low resolution screen capture of a webpage [23].”
• Malicious URL Detection using Machine Learning: A Survey (2017, methodology): “One of the earliest attempts at using visual features for this task was by computing the Earth Mover’s Distance between 2 images [56].”
• Cracking Classifiers for Evasion: A Case Study on the Google's Phishing Pages Filter (2016, methodology): “To minimize the impact of phishing attacks, a variety of methods have been proposed to detect phishing pages, involving machine learning [37][53][58] or other techniques [26][27][29][31][33][46][60][61].”