Distinguished Professor of Computer Science Pierre Baldi recently published a new paper, “Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy,” in the journal Gastroenterology. Working in a collaboration with Dr. William Karnes and his team in UCI’s Department of Medicine, Baldi and Ph.D. student Gregor Urban designed and trained deep convolutional neural networks (CNNs) to detect the polyps in colonoscopies in order to help doctors improve the adenoma detection rate (ADR). Using a set of 8,641 colonoscopy images containing 4,088 polyps, the trained CNN was able to identify polyps with a cross-validation accuracy of 96.4 percent. The system has the potential to increase ADR and reduce interval colorectal cancers.
Chancellor’s Professor of Computer Science Gene Tsudik will be giving a keynote talk titled “Mitigating Tension between Security and Safety in Low-End Embedded Devices”at next week’s 23rd European Symposium on Research in Computer Security (ESORICS 2018). The symposium, held in Barcelona, Spain at the Universitat Politecnica de Catalunya from Sept. 3-7, has become one of the leading European research events in computer security. According to its website, “The aim of ESORICS is to further the progress of research in computer security by establishing a European forum for bringing together researchers in this area, by promoting the exchange of ideas with system developers and by encouraging links with researchers in related areas.”
The symposium proceedings will be published by Springer in the Lecture Notes in Computer Science series (LNCS).
Researchers in many fields increasingly use cloud computing and storage for computationally intensive scientific tasks. However, how secure is data stored in public and private clouds? “Not secure enough,” according to Computer Science Professors Anton Burtsev and Gene Tsudik. Fortunately, the two were recently awarded a $1 million grant from the National Science Foundation (NSF) for their proposal: “Horizon: Secure Large-Scale Scientific Cloud Computing.” The Horizon project will be a part of the NSF Cybersecurity Innovation for Cyberinfrastructure (CICI) program created to “develop, deploy and integrate security solutions that benefit the scientific community by ensuring the integrity, resilience and reliability of the end-to-end scientific workflow.”
In early August, Computer Science Professor Ian Harris gave a presentation on social engineering at Black Hat USA, the massive annual security conference held in Las Vegas every summer, with over 15,000 attendees. His talk, “Catch Me, Yes We Can! Pwning Social Engineers Using Natural Language Processing Techniques in Real-Time,” attracted more than 900 attendees.
How can we provide users with an accurate impression of the terrain and sky when they experience in-flight simulation, or a detailed view of various decorative materials such as wood or metal when they explore virtual rooms? Assistant Professor of Computer Science Shuang Zhao has received a National Science Foundation (NSF) award of $500,000 to address this issue.
Social engineering poses a critical threat to information security, with cyberattackers recognizing that people are often more vulnerable to manipulation than a hardened computer system. “Social engineering targets the weakest link in the system, the human actors,” explains Computer Science Professor Ian Harris, who is working to address this issue as Principal Investigator of a grant from the National Science Foundation (NSF). The work, “Detecting Social Engineering Attacks Using Semantic Language Analysis,” falls under the NSF’s Secure and Trustworthy Cyberspace (SaTC) program, and Harris was awarded nearly $500,000 to study this growing threat.
According to the grant abstract, the project will “confront the problem of social engineering by developing automated approaches to detect social engineering attacks in real time and alert the victim before harm can occur.” Leveraging question answering and natural language understanding techniques, the goal is to identify conversational statements with malicious intent. “The attacker must always perform one of two dialog actions, either asking a question whose answer is private, or issuing a command to perform a forbidden operation.”
This work will also result in a large corpus of non-phishing social engineering attacks in the form of audio recordings and written transcripts, which will be made publicly available to support both further research into the topic and the development of courses on social engineering attacks.
Read more about Harris’ research on this topic on DarkReading.com.
— Shani Murray
Computer Science Ph.D. student Jihyun Park is the lead author on a paper that recently won the best paper award at the 11th International Conference on Educational Data Mining Conference (EDM 2018) in Buffalo, N.Y. The paper, “Understanding Student Procrastination via Mixture Models,” proposes a new approach based on statistical machine learning techniques that can extract and quantify patterns of procrastination observed from student clickstream data in online college courses. In particular, persistent procrastination over the duration of a course was found to be strongly predictive of poorer student outcomes, providing strong evidence that time management is critical for success in online courses.
Computer Science Professor Magda El Zarki and History Professor Pat Seed were thrilled to see their game, “Sankofa,” first on the list of Bronze Medal winners for the 2018 International Serious Play Awards. The awards recognize excellence in serious games designed for use in K-12 or higher education, and “Sankofa” is designed to teach history and anthropology in a fun and engaging way by bringing 19th-century Ghana to life through gameplay.
After entering a password, your regular computer keyboard might appear to look the same as always, but a new approach harvesting thermal energy can illuminate the recently pressed keys, revealing that keyboard-based password entry is even less secure than previously thought. Computer Science Ph.D. students Tyler Kaczmarek and Ercan Ozturk in the Donald Bren School of Information and Computer Sciences (ICS), working with Chancellor’s Professor of Computer Science Gene Tsudik, have exploited thermal residue from human fingertips to introduce a new insider attack — the Thermanator.