A Framework for Detection and Measurement of Phishing Attacks
-
Upload
noel-solis -
Category
Documents
-
view
112 -
download
1
description
Transcript of A Framework for Detection and Measurement of Phishing Attacks
A Framework for Detection and Measurementof Phishing AttacksReporter: Li, Fong Ruei
National Taiwan University of Science and Technology
Slide 1 (of 35)
Machine Learning and Bioinformatics Laboratory
ReferenceWorkshop On Rapid Malcode Proceedings of the 2007 ACM workshop on Recurring malcode Alexandria, Virginia, USA
SESSION: Threats Pages: 1 - 8 Year of Publication: 2007 ISBN:978-1-59593-886-2
Slide 2 (of 35)
Machine Learning and Bioinformatics Laboratory
OutlineIntroductionPhishing URL TypesModeling Phishing URLs
Feature AnalysisTraining With Features
Analysis and FindingsConclusion
Slide 3 (of 35)
Machine Learning and Bioinformatics Laboratory
INTRODUCTIONPhishing is form of identity theft
social engineering techniques sophisticated attack vectors
To harvest financial information from unsuspecting consumers.
Often a phisher tries to lure her victim into clicking a URL pointing to a rogue page.
Slide 4 (of 35)
Machine Learning and Bioinformatics Laboratory
PHISHING URL TYPESWe examined a black list of phishing URLs maintained by Google
This black list is used to provide phishing protection in Firefox
Slide 5 (of 35)
Machine Learning and Bioinformatics Laboratory
PHISHING URL TYPESThe prominent obfuscation techniques are:
Type I: Obfuscating the Host with an IP address
Type II: Obfuscating the Host with another Domain
Type III: Obfuscating with large host namesType IV: Domain unknown or misspelled
Slide 6 (of 35)
Machine Learning and Bioinformatics Laboratory
PHISHING URL TYPES
Slide 7 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSUsing logistic regression classifierFor training the model training black list and white list as followsWe use 1245 URLs from this list as our training
black listWe used a list of the top 1000 most popular
URLs as the basis of our training white list set
Slide 8 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSFeature Analysis
We categorize our features into four groups: Page Based Domain Based Type Based Word Based
Slide 9 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSPage Based :
a numeric value on a scale of [0,1] relative importance of a page within a set of web
pages
Slide 10 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSPage Based :
Slide 11 (of 35)
Page Rank distribution for the white list and black list URLs hostname
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSDomain Based
This category contains only one feature: whether or not the URL’s domain name can be found in
the White Domain Table.
Slide 12 (of 35)
Machine Learning and Bioinformatics Laboratory
Slide 13 (of 35)
MODELING PHISHING URLSDomain Based
51.2% of the white list URLs were present in the table
0.2% of the black list URLs were found in this table.
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSType Based
Type I URL Almost all non-phishing (white list) URLs in our training
data do not contain host obfuscation A significant portion of the phishing URLs are host
obfuscated with an IP address.
Type II URL portion of the black list URLs are Type II URLs.
Slide 14 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSType Based
Slide 15 (of 35)
Distribution of Type I and Type II URLs in the training data
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSType Based
Type III URL we determine the number of characters present after an
organization in the hostname
Slide 16 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSType Basednon-phishing URL
http://by124fd.bay124.hotmail.msn.com/cgi-bin/getmsg 0 characters after msn.com & before the path separator the maximum number noticed in a white list URL are 14
charactersType III phishing URLs
7.34 characters (on average) after the target before the path separator
a maximum of 63 characters
Slide 17 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSWord Based Features
Phishing URLs are found to contain several suggestive word tokens
login and signin are very often found in a phishing URLWe discarded all tokens with length < 5
containe several common URL parts such as http://, and www.
We discarded organization name tokens We further removed query parameters
Slide 18 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLS
Slide 19 (of 35)
Distribution of these features in our training set
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSTraining With Features
Our labeled data consisted of 2508 URLs 1245 were phishing URLs 1263 were benign URLs Phishing URLs were placed under the positive (true) class non-phishing ones were under the negative (false) class
66% of URLs were used for training and the remaining 34% were used as the test set
Slide 20 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSTo indicate the relative strength of each feature in identifying a Phishing URL we report the corresponding odds ratios, ecoefficient
Slide 21 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLS
Slide 22 (of 35)
Machine Learning and Bioinformatics Laboratory
MODELING PHISHING URLSEvaluation Result
We evaluated the trained model on the 34% test set split.
We performed our evaluation over multiple runs with randomized partitioning.
This evaluation gave us an average accuracy of 97.31% with True Positive Rate of 95.8 % False Positive Rate of 1.2%.
Slide 23 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSWe collected several million URLs from August 20th to August 31 2006
The data consisted of two main components , unique URLs which are visited each dayconsecutive look up requests to these URLs
Slide 24 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Phishing URLs per day.The average number of phishing URLs which have been visited from Google’s toolbar in a day.
we find that on average there are 777 URL phishing attacks in a day5073 viewers to a phishing page
Slide 25 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Phishing URLs per day.
Slide 26 (of 35)
the distribution of phishing attacks on each day of our study.
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Phishing URLs per day.
Slide 27 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Phishing URLs per day.
Slide 28 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Potential Phishing Victims per day.Determine how many users interact with a phishing page
A user that has any interaction at a site classified as phishing is regarded as a potential phishing victim.
Slide 29 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Potential Phishing Victims per day.Based on the number of users who view phishing pages in a day, we further can infer Potential Success Rate of a phisher as follows:
Slide 30 (of 35)
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSAverage Potential Phishing Victims per day.
Slide 31 (of 35)
the distribution of phishing attacks on each day of our study.
Machine Learning and Bioinformatics Laboratory
ANALYSIS AND FINDINGSDistribution of Phishing by Organization
Slide 32 (of 35)
Machine Learning and Bioinformatics Laboratory
Slide 33 (of 35)
ANALYSIS AND FINDINGSGeographical Distribution of Phishing.
To determine country that hosts a particular phishing URL, we used Google’s IP to Geo-Location infrastructure.
Machine Learning and Bioinformatics Laboratory
Anti-Phishing Tools
Slide 34 (of 35)
Machine Learning and Bioinformatics Laboratory
CONCLUSIONWe use our features in a logistic regression classifier that achieves a very high accuracy.
One of the major contributions of this work is a large scale measurement study conducted on Google Toolbar URLs
On average we found around 777 unique phishing pages per day and on average 8.24% of the number users who view phishing pages are potential phishing victims
Slide 35 (of 35)