The Internet has created vast opportunities for interacting with strangers. The interactions can be fun, informative, and even profitable. However, there is also risk involved. Will a seller at eBay ship the product in time? Is the advice from a self-proclaimed expert at Epinion.com trustworthy? Does a product at Amazon.com have high quality as described?
To address these problems, one of the most ancient mechanisms in the history of human society, word-of-mouth, is gaining new significance in the cyber space, where it is called reputation system. A reputation system collects evidence about the properties of online items, analyzes and aggregates the evidence, and disseminates the aggregated results. Here, the item can be people (e.g. in eBay), a product (e.g. in Amazon product rating), or a piece of digital information (e.g. video clip at YouTube). The aggregated results are called reputation scores. Most commercial systems collect user feedback/ratings as evidence. This type of system is referred to as feedback-based reputation system or rating system.
With the rapid development of online reputation systems, the incentive to manipulate and mislead the reputation systems is growing fast.
(1) Detection of Collaborated Cheating Behaviors in Online Reputation Systems
The power of one online identity is limited. Powerful attacks are usually launched by multiple colluded malicious user IDs. Can we take advantage of this collusion feature to identify the malicious users?
(2) Defending Reputation Manipulation from Reputation Boosting Companies
Recently, many "rating companies" have generated or collected a large affiliate network of online user IDs, to provide rating manipulation services for different customers. For just $9.99, a company named "IncreaseYouTubeViews.com" can provide 30 "I like" ratings or 30 real user comments to your video clips on YouTube. It indicates that in the online market, it is highly possible that the reputation scores of many items are manipulated by one company. How to identify these items?
Towards Trustworthy Neural-controlled Artificial Legs using High-Performance Embedded Computers
There are over 32 million amputees worldwide whose life are severely impacted. A continued need exists to provide this large and growing population of amputees with the best care and return of function possible. The quality of life of leg amputees can be improved dramatically by using a cyber physical system (CPS) that controls artificial legs based on neural signals representing amputees’ intended movements. The key to the CPS system is the neural-machine interface (NMI) that senses electromyographic (EMG) signals to make control decisions.
In this design, the security issue is very important, since any wrong decisions may lead to falls of the amputees, which is very dangerous. We apply trust model in this application to evaluate the reliability of the EMG sensors in real time, help the NMI system to make sensor removal and entering decisions and ensure the security of the entire NMI system.
Some experiment videos are available at
Social network facilitated security solutions
Investigating effect of App sale boosting services
With the wide spread of smartphones and tablet computers, the sale of mobile applications is experiencing an overwhelming growth. In June 2012, Apple's app store has hit the 30 billion downloads milestone. Accompanied with this, dozens of other companies, such as Google, Amazon, blackberry, have also opened their own app markets and are achieving big success. The app markets have provided unique opportunities for big companies, small businesses, and even independent developers to gain enormous profit.
With the big success of the mobile application (app) sales, how to make profit from app markets has attracted wide attention. One important question is how to accurately identify potential customers.
Detection of cheating behaviors in cyber competition
Cyber Competition has been recognized as an efficient way to facilitate research and education in cyber security field. We have discovered that the participants (i.e. players) in cyber competitions can cheat in order to gain a higher rank or collect more prizes. In this work, we use data collected from the CANT competition to analyze such cheating behaviors and propose to build a competition social network to detect cheating behaviors in cyber competitions.