<
Content Page >
All Contents | Last updated on 30Oct2011, Sunday
Anti-Phishing Tests - October 2011
Note: These tests are not conducted by Chimalogy but by Av-Comparatives.
Test procedure
In our common test scenario we simulate a user which relies on the Anti-Phishing protection provided by its security product while browsing the web (and/or checking his web mail
e-mail account, i.e. anti-spam features are not considered as they are not the scope of this test).
The test was done using Windows XP SP3 and Internet Explorer 7 (without the build-in phishing blocker in order to get browser-independent results) under VMware. All security products were tested using default settings. All products were tested in parallel at the same time on the same URLs.
Tested products (19)
• AVG Internet Security 2011
• Avira Premium Security Suite 10.1
• Bitdefender Internet Security 2012
• Bullguard Internet Security 10.0
• eScan Internet Security 11.0
• ESET Smart Security 4.2
• F-Secure Internet Security 2011
• G DATA Internet Security 2012
• K7 Total Security 11.1
• Kaspersky Internet Security 2012
• McAfee Total Protection 2011
• Panda Internet Security 2012
• PC Tools Internet Security 2011
• Qihoo 360 Internet Security 2.0
• Quick Heal Internet Security 2011
• Symantec Norton Internet Security 2011
• Trend Micro Titanium Internet Security 2012
• Trustport Internet Security 2012
• Webroot Internet Security Complete 7.0
We were asked to test also Avast, Microsoft (Security Essentials) and Sophos. But Avast and Sophos do not have any Anti-Phishing feature, except blocking phishing mails with their spam filter. Microsoft does block phishing websites in its browser (Microsoft Internet Explorer) and its spam filter (e.g. Microsoft Outlook); Security Essentials is not designed for this task.
ESET doesn’t have a dedicated module for anti-phishing protection yet. However, phishing sites are currently blocked along with other potentially harmful sites (like in most other products), employing various technologies such as web filter, parental control and detection of suspicious content aimed at the phishing landing pages themselves.
Qihoo is mainly targeted to block Chinese phishing sites.
Test Set
Phishing URLs taken out from spammed prevalent phishing mails and/or collected by our crawler and researchers. All phishing URLs had to be active/online and attempt to get personal information. Duplicate phishing URLs, as well as phishing campaigns hosted on same server/IP were removed.
All test cases were manually re-verified after the test. At the end, only 697 different and valid Phishing URLs remained. The phishing campaigns targeted various types of personal data. Among those were (in the following order) phishing attempts to gather e.g. login credentials etc. for: PayPal, eBay, Online Banking & Credit cards, Social networks, Online Games,
E-mail accounts and other online services.
Below you see the percentages of blocked phishing websites (size of test set: 697 phishing URLs). Please take into consideration the false alarm rates (on next pages) when looking at the below results (products with false alarms are marked with an asterisk).
(Click on the image for a clearer view) (Click on the image for a clearer view)
Anti-Phishing “False Alarm” Test
For the Anti-Phishing False Alarm Test we selected 1000 legitimate banking sites (all of them using HTTPS) from all over the world and checked if those legitimate online banking sites were blocked by the various security products.
Wrongly blocking such sites is a serious mistake. From the tested 19 products, only two (K7 and Webroot) had false alarms on the tested 1000 legitimate online banking sites:
K7 (1 false alarm):
• Bank Islam from Malaysia
Webroot (12 false alarms):
• AB Finance Bank from Russia
• Bank of Kenya from Kenya
• BLC Bank from Lebanon
• Dawia Next Bank from Japan
• Dhanlaxmi Bank from India
• Habib Bank Zuricj from UAE
• ING NL from Netherlands
• Marfin Laiki Bank from Cyprus
• Refah Bank from Israel
• Saxo eBank from Sweden
• ScotiaBank from Mexico
The discovered false alarms have been reported to the respective vendors and are now no longer blocked.
Which products support which browser?
Most browsers already include their own Anti-Phishing technologies. Nevertheless, due to the complexity and amount of social engineered scams and phishing attempts, it is recommended to make use of the Anti-Phishing features provided by security products too.
Practically all security products support the main stream browsers (i.e. Internet Explorer and Firefox), while only few support also other browsers. Due to that, it may be safer to make use of a supported browser to perform e.g. online banking.
(Click on the image for a clearer view)
Passed - Green (Supports all of the 5 commonly used browsers)
Neutral - Orange (Supports at least 3 commmonly used browsers)
Failed - Red (Supports less than 3 commonly used browsers)
Passed:
AVIRA Premium Security Suite
Bitdefender Internet Security
ESET Smart Security
G DATA Internet Security
Kaspersky Internet Security
Panda Internet Security
Trend Micro Titanium Internet Security
Neutral:
PC Tools Internet Security
Symantec Norton Internet Security
Failed:
AVG Internet Security
Bullguard Internet Security
eScan Internet Security
F-Secure Internet Security
K7 Total Security
McAfee Total Protection
Qihoo 360 Internet Security
Quick Heal Internet Security
Trustport Internet Security
Webroot Internet Security
Conclusion:
Top 7 that passes the overall tests.
1) 94.4% - Bitdefender
2) 92.8% -Trend Micro
3) 89.1% - AVIRA
4) 88.5% - G DATA
5) 85.1% - Kaspersky
6) 84.5% - Panda
7) 78.0% - ESET
Note:
I'm suprised that ESET managed to pass since it doesn’t have a dedicated module for anti-phishing protection yet but phishing sites is currently blocked along with other potentially harmful sites.
Chance of getting phished = Your Common Sense + Browsers + Security Products
In fact, this test seems pretty useless(except is show where is your anti-virus standing) to me as common sense play the largest part in this.
My advise is to always keep your software like browser and security product up-to-date and never open/click on any suspicious links from unknown and known person. Always double check with the person before clicking the link and is possible, type the link in your browser instead of clicking it.
Reference:
<
Back To Content Page >
0 comments:
Post a Comment