Detecting Fraudulent Skype Users via Machine Learning

#artificialintelligence

As part of my Data Science class with General Assembly, we each gave a presentation about a real-world application of data science. My talk was about using machine learning to detect fraud on Skype, and was based upon an excellent paper by Microsoft Research published in November 2013. Although Skype already had measures in place to detect fraud (e.g., credit card fraud, spam instant messages), the research team's goal was to improve the detection of "stealthy fraudulent users" that evade Skype's defenses for a prolonged period. They built a machine learning classifier that flagged potentially fraudulent users, and was able to detect 68% of these users with a false positive rate of 5%. The novelty in their approach was the fusing of disparate data types (profile information, Skype product usage, and Skype social activity) into a single classifier.


Generic OS X Malware Detection Method Explained

#artificialintelligence

When it comes to detecting OS X malware, the future may not be rooted in machine learning algorithms, but patterns and heatmap visualization, a researcher posits. In an academic paper published by Virus Bulletin on Monday, Vincent Van Mieghem, a former student at the Delft University of Technology in the Netherlands, describes how a recurring pattern he observed in OS X system calls can be used to indicate the presence of malware. Van Mieghem wrote the paper, "Behavioral Detection and Prevention of Malware on OS X," (.PDF) while interning at Fox-IT but has since moved on to PricewaterhouseCoopers' cybersecurity division. By the numbers, the detection method Van Mieghem concocted is a success; it detected infections from 100 percent of malware samples found on OS X systems at the time. The method apparently leaves little room for error too; it resulted in a scant 0 percent to 20 percent false positive rate, depending on the user, according to the paper.


Joint Attention and Brain Functional Connectivity in Infants and Toddlers Cerebral Cortex

#artificialintelligence

Initiating joint attention (IJA), the behavioral instigation of coordinated focus of 2 people on an object, emerges over the first 2 years of life and supports social-communicative functioning related to the healthy development of aspects of language, empathy, and theory of mind. Deficits in IJA provide strong early indicators for autism spectrum disorder, and therapies targeting joint attention have shown tremendous promise. However, the brain systems underlying IJA in early childhood are poorly understood, due in part to significant methodological challenges in imaging localized brain function that supports social behaviors during the first 2 years of life. Herein, we show that the functional organization of the brain is intimately related to the emergence of IJA using functional connectivity magnetic resonance imaging and dimensional behavioral assessments in a large semilongitudinal cohort of infants and toddlers. In particular, though functional connections spanning the brain are involved in IJA, the strongest brain-behavior associations cluster within connections between a small subset of functional brain networks; namely between the visual network and dorsal attention network and between the visual network and posterior cingulate aspects of the default mode network. These observations mark the earliest known description of how functional brain systems underlie a burgeoning fundamental social behavior, may help improve the design of targeted therapies for neurodevelopmental disorders, and, more generally, elucidate physiological mechanisms essential to healthy social behavior development. The emergence of joint attention (JA), the coordinated orienting of 2 people toward an object or event, occurs during the first 2 years of life, arguably the most dynamic and important period of early child development (Scaife and Bruner 1975). It is theorized that engaging in JA lays the foundation for prosocial cooperative behavior, from basic social-communicative functioning and language development (Premack 2004) to sophisticated forms of empathy (Mundy and Jarrold 2010) and theory of mind (Adolphs 2003). In fact, early exhibition of joint attention is strongly associated with later language ability (Morales et al. 2000; Mundy et al. 2007), and atypical development of the initiation of joint attention (IJA) is strongly indicative of autism spectrum disorder (ASD) (Bruinsma et al. 2004). The neural substrates underlying IJA in early childhood are poorly understood (Barak and Feng 2016), due in part to significant methodological challenges in imaging localized brain function that supports social behaviors in children during the first 2 years of life.


Sophos Adds Advanced Machine Learning to Its Next-Generation Endpoint Protection Portfolio with Acquisition of Invincea

#artificialintelligence

Sophos (LSE: SOPH), a global leader in network and endpoint security, today announced it has entered into an agreement to acquire Invincea, a visionary provider of next-generation malware protection. Invincea's endpoint security portfolio is designed to detect and prevent unknown malware and sophisticated attacks via its patented deep learning neural-network algorithms. It has been consistently ranked as among the best performing machine learning, signature-less next-generation endpoint technologies in third-party testing and rated highly both for high detection and low false-positive rates. Headquartered in Fairfax, Va., Invincea was founded by chief executive officer Anup Ghosh to address the rapidly growing zero-day security threat from nation states, cyber criminals and rogue actors. Invincea's flagship product X by Invincea uses deep learning neural networks and behavioral monitoring to detect previously unseen malware and stops attacks before damage occurs.


A Comparison between Deep Neural Nets and Kernel Acoustic Models for Speech Recognition

#artificialintelligence

We study large-scale kernel methods for acoustic modeling and compare to DNNs on performance metrics related to both acoustic modeling and recognition. Measuring perplexity and frame-level classification accuracy, kernel-based acoustic models are as effective as their DNN counterparts. However, on token-error-rates DNN models can be significantly better. We have discovered that this might be attributed to DNN's unique strength in reducing both the perplexity and the entropy of the predicted posterior probabilities. Motivated by our findings, we propose a new technique, entropy regularized perplexity, for model selection.