UO's Lowd aims to take on spammers with young investigator award

Don't let the title of a new project by UO computer scientist Daniel Lowd mess up your brain. It's all about detecting malicious behavior such as spam on social networks or fake reviews on various Internet sites.

Lowd has won an Army Research Office's Young Investigator Award to pursue his proposal for "Inferring Trustworthiness and Deceit in Adversarial Relational Models." The award, which provides $360,000 over three years, is given to promising academic scientists who have received doctorates or equivalent degrees within the last five years.

Lowd will seek to separate honest and dishonest activity by exploiting a relational structure often used by spammers. For example, he says, social network spammers tend to be friends with other spammers, and users who have posted one fake review tend to post several more. Thus, evidence of malicious behavior can be used to find more malicious behavior elsewhere in the network.

He'll use a variety of techniques, including game theory. In previous work, Lowd, an assistant professor in the Department of Computer and Information Science, helped pioneer the area of adversarial machine learning, demonstrating that spam filters and similar systems are vulnerable to attacks, even when the filter's parameters are kept secret.

Lowd, a member of the UO's newly launched Center for Cyber Security and Privacy, joined the UO in 2009 while completing his doctorate in computer science from the University of Washington. He received his doctorate in 2010.