Science of Security and Privacy 2022 Annual Report
Science of Security and Privacy 2022 Annual Report
The attached Science of Security and Privacy 2022 Annual Report highlights the progress and accomplishments of this initiative.
Second News Item
Second News Item
This is another news item.
Submitted by Matthew Warner
on
Trustworthy AI
|
Recent years have seen an astounding growth in deployment of AI systems in critical domains such as autonomous vehicles, criminal justice, healthcare, hiring, housing, human resource management, law enforcement, and public safety, where decisions taken by AI agents directly impact human lives. Consequently, there is an increasing concern if these decisions can be trusted to be correct, reliable, fair, and safe, especially under adversarial attacks. |
Demo: Securing Automotive Zonal Architectures with Named Data Networking
Data-centric architectures are a candidate for in-vehicle communication. They add naming standardization, data provenance, security, and improve interoperability between different ECUs and networks. In this demo, We demonstrate the feasibility and advantages of data-centric architectures through named Data Networking (NDN). We deploy a bench-top testbed using Raspberry PIs to demonstrate NDNs value in a segmented network. We replay real CAN data over a CAN hat to NDN gateways and demonstrate how that data can be forwarded between segments using NDN.
Identifying Online Misbehavior
Online misbehavior such as stalking, doxxing, inappropriate messages, is prevalent on mobile applications. We observed that app users report such incidents in the app reviews. Victims of such incidents are intimidated and feel unsafe on the apps. App developers and the app distribution stores should be aware of such incidents, and take actions to rectify the misuse problem. We explored multiple Natural Language Processing (NLP) techniques to identify such incidents from the app reviews data set. Our best model achieved 80.63% precision at 73.48% recall.
DREAD: Dark web Retrieval, Examination, And Detection
The Dark Web is infamous for abetting criminal activity by providing some level of anonymity for its users through the use of the Tor Network. The sale of banned substances, weapons, and illegal services is frequent on the dark web, creating a significant problem for law enforcement at every level. To combat this issue, we have developed DREAD: a Dark Web search engine designed to identify banned substances, weapons, and illicit activity on the Dark Web.
Applying Software Security Growth Model to Web Browser Software
Traditional software reliability growth models limit consideration of the software reliability metrics to software testing and defect data only. In contrast, vulnerability discovery models consider post-release vulnerability data and apply software reliability growth models to future software versions. However, vulnerability data are sparse and are a subset of all undiscovered vulnerabilities, resulting in overestimating the vulnerability prediction in the next software release.
Toward Automatic Detection of Cloud Server Security Vulnerabilities
Cloud systems have been widely adopted in many real world production applications. Thus, security vulnerabilities in those cloud systems can cause serious widespread impact. Although previous intrusion detection systems can detect security attacks, understanding the underlying software defects that cause those security vulnerabilities is little studied. In this work, we conduct a systematic study over 109 software security vulnerabilities in 13 popular cloud server systems.