- Fukushima Report: Robot Lifts Melted Fuel
- TÜV, Nozomi Ink Partnership Pact
- Pangea Patches Bypass Vulnerability
- Fuji Fixes FRENIC Devices
- ARC: Safety and Profitability Work Together
- Public Needs to Know About Chem Releases: Judge
- Robot Testing Radioactive Fuel at Fukushima
- Siemens Fixes CP1604, CP1616 Holes
- Siemens has Upgrade for Intel AMT
- Siemens Fixes Hole in SIMATIC S7-300 CPU
- Siemens has Licensing Software Fix for SICAM 230
- Siemens Fixes Ethernet Communication Module, Relays
- OSIsoft has Update for PI Vision Hole
- First Responders Test Technology
- Manufacturing Targeted in Hack Attack
- Siemens Fixes SICAM A8000 RTU Series Hole
Chemical Safety Incidents
Looking to Hike Online Social Networks Security
Wednesday, May 4, 2016 @ 09:05 AM gHale
Online social networks are a hotbed of potential fraud and attacks and there is a movement to figure out a way to detect those crimes.
University of Arkansas computer science and engineering professor Xintao Wu has received $348,758 from the National Science Foundation as a part of a four-year project called, “Online Social Network Fraud and Attack Research and Identification.”
RELATED STORIES
Working to Quash Quantum Computer Threat
DHS’ Risk Analysis Tool Licensed
Phishing Attacks Prevented by SCAM
Rapid Address Change Aids Security
Wu is collaborating with Jun Li, computer science professor at the University of Oregon, and Aidong Lu, computer science professor at the University of North Carolina at Charlotte.
Online social networks such as Facebook, Instagram and Pinterest are susceptible to various types of fraud and attacks, like account impersonation, viral marketing and Sybil attacks, which forge identities across networks to subvert systems. Existing security approaches usually target a specific type of fraud or attack and often fall short of detecting more complex attacks, such as collusive attacks that involve fraudulent network accounts or dynamic attacks that encompass multiple attack phases over time.
Wu’s project models online social network threats to identify, evaluate and analyze specific forms of fraud and attacks. The researchers will develop a defense against attack by using a spectral-analysis-based framework for detection. The system will automatically detect dynamic attacks and transform attack characteristics into distinctive visual patterns. The system will include interactive mechanisms that allow analysts to incorporate domain knowledge to handle attacks.
The research team will build a simulation framework to evaluate the detection approaches against different types of network attacks. Users can plug in different datasets to evaluate and compare different detection approaches.
The research team plans to develop a course focused on online social network attacks and defenses. Wu said the course has the potential to be offered across many institutions. To increase public-security awareness, the team also plans to develop tutorial courses on typical network attacks and how to defend against them. They intend to offer these courses at popular public events and in freshman classes. The research team will broadly disseminate their results, tools, software and documents to the research community, information-technology industries and to online social network companies.
Leave a Reply
You must be logged in to post a comment.