2017 · 16 citations · 16 references
Social SciencesCommunicationComputer ScienceComputer SecuritySocial Media AnalyticsText AnalyticsSocial MediaUser Behavior AnalysisVirtual CommunitiesCommunication PatternsIndicator ExtractionSocial Media Threats
Hackers make extensive use of online communities, sharing knowledge, tools, as well as performing coordination and recruitment activities. In order to detect such behaviors, this paper proposes a set of indicators which analyze online communication patterns, including technical discussions, expression of positive and negative sentiments and threats, recruitment activities, and user profiling. The indicators are processing streaming social media and search for online behaviors and communication patterns characteristic of hackers with different motivations and skills. Our initial evaluation of indicators using twitter data shows that there is a significant variation in indicator values across different types of hackers. For example, hackers with higher level skills tend to use technical topics in their conversation more often than hackers with lower skills, whereas hackers motivated by profit and ideology tend to express recruitment language more often than attackers motivated by revenge and prestige. These results support our hypothesis that detection of hacking behaviors on social media needs to take into account the differences in intentions, motivations, and skills of different types of hackers.
16
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
Richard Socher, Alex Perelygin, Jean Y. Wu et al. · 2013 · 6.8K citations · Full text
An analysis of underground forums
Marti Motoyama, Damon McCoy, Kirill Levchenko et al. · 2011 · 310 citations
The simple economics of cybercrimes
Nir Kshetri · IEEE Security & Privacy · 2006 · 140 citations