Analytics // Security Monitoring
8/24/2012
08:56 PM
Connect Directly
RSS
E-Mail
50%
50%

Application Detects Social Network Spam, Malware

Using the social context of posts, researchers from UC Riverside create prototype Facebook app that detects social malware with 97 percent accuracy

Social networks are both a boon and bane for online criminals: People using the networks tend to trust messages sent by "friends" and other users to whom they are connected, making social engineering that much more effective. On the other hand, the networks are gated communities, where security policies and technologies can radically change the attack landscape.

MyPageKeeper, a project designed and created by computer scientists at the University of California, Riverside, does just that. Created as a Facebook application, the program searches the news feeds of its subscribers every two hours looking for suspected social malware and scams, collectively referred to as "socware" by the researchers. When it finds a suspect post, it leaves a comment indicating that the item is likely a scam or malware.

The group was able to correlate posts that were duplicated across MyPageKeeper's user base -- on sign of spamminess -- and posts that contained links to suspect Internet addresses, accurately identifying socware in 97% of cases.

"The core technology underlying this app is how to efficiently see what posts are pointing to malicious Web pages," says Harsha V. Madhyastha, an assistant professor in computer science and engineering at UC Riverside and a co-author of a paper on the project presented at the USENIX Security Conference earlier this month. "We identify what is suspicious without crawling every link in every post. To minimize our costs, we identify what is bad and what is good just based on what we call the social context -- the information included in the post and the URLs."

Socware is prevalent on social networks. Almost half of all users had encountered a scam or malware on Facebook in the prior four months, according to the researchers. In addition, socware is tailored to take advantage of the way each service works. For example, likejacking -- referred to by the researchers as "like as a service" -- is a method for fooling users into inadvertently raising the reputation of Facebook applications or posts.

"Typically per day, we find 50 unique URLs, and those are spread across hundreds of users," Madhyastha says. "And this is from our subset of Facebook users that are out there."

[ Recent widespread spam runs posing as convincing-looking email messages from LinkedIn, Facebook, ADP, American Express, US Airways, the U.S. Postal Service, UPS, and several other high-profile organizations are all part of a single, orchestrated attack campaign. See Series Of Convincing Spam Runs Part Of One Massive Advanced Attack Campaign.]

The project suggests ways of curtailing the pollution of spam, malicious apps, and fraudulent links on social networking sites, such as Facebook and Twitter. Earlier this month, Facebook estimated that 8.7 percent of the accounts on its site are fake in some way.

The techniques could also help companies protect their employees on social networks, identifying potentially fraudulent posts. The researchers are also investigating ways of rolling back posts and tweets that a person may regret posting or that violate company policy.

Other technologies have struggled with solving the problem of malicious activities on social networks, says Michael Sutton, vice president of security research at Web security firm Zscaler. Attackers have already adapted to defensive techniques, such as URL filtering and antivirus.

"The old ways of doing security are really struggling with things such as social networks," he says.

The problem with the UC Riverside approach, however, is that it is specific to each social network. The provider would have to create an app for each provider's application programming interface.

Yet, socware exists almost entirely in the social networking world, and so it makes most sense to solve the problem there, says Adam Wosotowsky, messaging data architect with McAfee.

"This is a good technique -- a version 1.0 to be able to sure, but the direction, I think, is the right direction," he says. "If you want to protect in the realm of the social networking service, you have to be in the realm of the social networking service."

Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Candace
50%
50%
Candace,
User Rank: Apprentice
8/25/2012 | 6:01:59 AM
re: Application Detects Social Network Spam, Malware
Kudos to the UC Riverside team for taking on social spam. Their findings underscore how the spam problem is growing and why site owners should consider defensive technologies against it. Unfortunately, social spam creates real costs in-ávarious forms including lost users, tarnished reputation, skewed analytics, and decreased ad revenue. Automated moderation, leveraging analysis of context, URLs, and other features, is definitely the way to go.
Register for Dark Reading Newsletters
White Papers
Flash Poll
Current Issue
Cartoon
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2013-6117
Published: 2014-07-11
Dahua DVR 2.608.0000.0 and 2.608.GV00.0 allows remote attackers to bypass authentication and obtain sensitive information including user credentials, change user passwords, clear log files, and perform other actions via a request to TCP port 37777.

CVE-2014-0174
Published: 2014-07-11
Cumin (aka MRG Management Console), as used in Red Hat Enterprise MRG 2.5, does not include the HTTPOnly flag in a Set-Cookie header for the session cookie, which makes it easier for remote attackers to obtain potentially sensitive information via script access to this cookie.

CVE-2014-3485
Published: 2014-07-11
The REST API in the ovirt-engine in oVirt, as used in Red Hat Enterprise Virtualization (rhevm) 3.4, allows remote authenticated users to read arbitrary files and have other unspecified impact via unknown vectors, related to an XML External Entity (XXE) issue.

CVE-2014-3499
Published: 2014-07-11
Docker 1.0.0 uses world-readable and world-writable permissions on the management socket, which allows local users to gain privileges via unspecified vectors.

CVE-2014-3503
Published: 2014-07-11
Apache Syncope 1.1.x before 1.1.8 uses weak random values to generate passwords, which makes it easier for remote attackers to guess the password via a brute force attack.

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
Marilyn Cohodas and her guests look at the evolving nature of the relationship between CIO and CSO.