BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Yuval Elovici

Yuval Elovici appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
Cryptography and SecurityMachine LearningSocial and Information Networkscs.CYphysics.soc-phArtificial IntelligenceComputer VisionDistributed, Parallel, and Cluster ComputingSoftware EngineeringApplicationsDigital Librarieseess.IVHuman-Computer InteractionMultiagent SystemsNetworking and Internet ArchitecturePopulations and Evolution

Catalog footprint

What is connected

52works
16topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

52 published item(s)

preprint2026arXiv

AgentGuardian: Learning Access Control Policies to Govern AI Agent Behavior

Artificial intelligence (AI) agents are increasingly used in a variety of domains to automate tasks, interact with users, and make decisions based on data inputs. Ensuring that AI agents perform only authorized actions and handle inputs appropriately is essential for maintaining system integrity and preventing misuse. In this study, we introduce the AgentGuardian, a novel security framework that governs and protects AI agent operations by enforcing context-aware access-control policies. During a controlled staging phase, the framework monitors execution traces to learn legitimate agent behaviors and input patterns. From this phase, it derives adaptive policies that regulate tool calls made by the agent, guided by both real-time input context and the control flow dependencies of multi-step agent actions. Evaluation across two real-world AI agent applications demonstrates that AgentGuardian effectively detects malicious or misleading inputs while preserving normal agent functionality. Moreover, its control-flow-based governance mechanism mitigates hallucination-driven errors and other orchestration-level malfunctions.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Peacock: UEFI Firmware Runtime Observability Layer for Detection and Response

Modern computing platforms rely on the Unified Extensible Firmware Interface (UEFI) to initialize hardware and coordinate the transition to the operating system. Because this execution environment operates with high privileges and persists across reboots, it has increasingly become a target for advanced threats, including bootkits documented in real systems. Existing protections, including Secure Boot and static signature verification, are insufficient against adversaries who exploit runtime behavior or manipulate firmware components after signature checks have completed. In contrast to operating system (OS) environments, where mature tools provide dynamic inspection and incident response, the pre-OS stage lacks practical mechanisms for real-time visibility and threat detection. We present Peacock, a modular framework that introduces integrity-assured monitoring and remote verification for the UEFI boot process. Peacock consists of three components: (i) a UEFI-based agent that records Boot and Runtime Service activity with cryptographic protection against tampering; (ii) a cross-platform OS Agent that extracts the recorded measurements and produces a verifiable attestation bundle using hardware-backed guarantees from the platform's trusted module; and (iii) a Peacock Server that verifies attestation results and exports structured telemetry for enterprise detection. Our evaluation shows that Peacock reliably detects multiple real-world UEFI bootkits, including Glupteba, BlackLotus, LoJax, and MosaicRegressor. Taken together, these results indicate that Peacock provides practical visibility and verification capabilities within the firmware layer, addressing threats that bypass traditional OS-level security mechanisms.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SAFEdit: Does Multi-Agent Decomposition Resolve the Reliability Challenges of Instructed Code Editing?

Instructed code editing is a significant challenge for large language models (LLMs). On the EditBench benchmark, 39 of 40 evaluated models obtain a task success rate (TSR) below 60 percent, highlighting a gap between general code generation and the ability to perform instruction-driven editing under executable test constraints. To address this, we propose SAFEdit, a multi-agent framework for instructed code editing that decomposes the editing process into specialized roles to improve reliability and reduce unintended code changes. A Planner Agent produces an explicit, visibility-aware edit plan, an Editor Agent applies minimal, literal code modifications, and a Verifier Agent executes real test runs. When tests fail, SAFEdit uses a Failure Abstraction Layer (FAL) to transform raw test logs into structured diagnostic feedback, which is fed back to the Editor to support iterative refinement. We compare SAFEdit against both prior single-model results reported for EditBench and an implemented ReAct single-agent baseline under the same evaluation conditions. We used EditBench to evaluate SAFEdit on 445 code editing instances in five languages (English, Polish, Spanish, Chinese, and Russian) under varying spatial context variants. SAFEdit achieved 68.6 percent TSR, outperforming the single-model baseline by 3.8 percentage points and the ReAct single-agent baseline by 8.6 percentage points. The iterative refinement loop was found to contribute 17.4 percentage points to SAFEdit's overall success rate. SAFEdit's automated error analysis further indicates a reduction in instruction-level hallucinations compared to single-agent approaches, providing an additional framework component for interpreting failures beyond pass or fail outcomes.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SecMate: Multi-Agent Adaptive Cybersecurity Troubleshooting with Tri-Context Personalization

Recent advances in large language models and agentic frameworks have enabled virtual customer assistants (VCAs) for complex support. We present SecMate, a multi-agent VCA for cybersecurity troubleshooting that integrates device, user, and service specificity from conversational and device-level signals. Device specificity is provided by a lightweight local diagnostic utility, while user specificity relies on implicit proficiency inference and profile-aware troubleshooting. Service specificity is achieved through a proactive, context-aware recommender. We evaluate SecMate in a controlled study with 144 participants and 711 conversations. Device-level evidence increased correct resolutions from about 50% to over 90% relative to an LLM-only baseline, while step-by-step guidance improved pleasantness and reduced user burden. The recommender achieved high relevance (MRR@1=0.75), and participants showed strong willingness to substitute human IT support at costs well below human benchmarks. We release the full code base and a richly annotated dataset to support reproducible research on adaptive VCAs.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs

As organizations increasingly deploy LLMs in sensitive domains such as legal, financial, and medical settings, ensuring alignment with internal organizational policies has become a priority. Existing content moderation frameworks remain largely confined to the safety domain and lack the robustness to capture nuanced organizational policies. LLM-as-a-judge and fine-tuning approaches, though flexible, introduce significant latency and training cost. To address these limitations, we frame policy violation detection as an out-of-distribution (OOD) problem in the model's activation space. We propose a training-free method that operates directly on the LLM internal representations, leveraging prior evidence that decision-relevant information is encoded within them. Inspired by whitening techniques, we apply a linear transformation to decorrelate and standardize the model's hidden activations, and use the Euclidean norm in this transformed space as a compliance score for detecting policy violations. Our method requires only the policy text and a small number of illustrative samples, making it lightweight and easily deployable. We extensively evaluate our method across multiple LLMs and challenging policy benchmarks, achieving 86.0% F1 score while outperforming fine-tuned baselines by up to 9.1 points and LLM-as-a-judge by 16 points, with significantly lower computational cost. Code is available at: https://github.com/FujitsuResearch/LLM-policy-violation-detection

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Adversarial Mask: Real-World Universal Adversarial Attack on Face Recognition Model

Deep learning-based facial recognition (FR) models have demonstrated state-of-the-art performance in the past few years, even when wearing protective medical face masks became commonplace during the COVID-19 pandemic. Given the outstanding performance of these models, the machine learning research community has shown increasing interest in challenging their robustness. Initially, researchers presented adversarial attacks in the digital domain, and later the attacks were transferred to the physical domain. However, in many cases, attacks in the physical domain are conspicuous, and thus may raise suspicion in real-world environments (e.g., airports). In this paper, we propose Adversarial Mask, a physical universal adversarial perturbation (UAP) against state-of-the-art FR models that is applied on face masks in the form of a carefully crafted pattern. In our experiments, we examined the transferability of our adversarial mask to a wide range of FR model architectures and datasets. In addition, we validated our adversarial mask's effectiveness in real-world experiments (CCTV use case) by printing the adversarial pattern on a fabric face mask. In these experiments, the FR system was only able to identify 3.34% of the participants wearing the mask (compared to a minimum of 83.34% with other evaluated masks). A demo of our experiments can be found at: https://youtu.be/_TXkDO5z11w.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

AnoMili: Spoofing Prevention and Explainable Anomaly Detection for the 1553 Military Avionic Bus

MIL-STD-1553, a standard that defines a communication bus for interconnected devices, is widely used in military and aerospace avionic platforms. Due to its lack of security mechanisms, MIL-STD-1553 is exposed to cyber threats. The methods previously proposed to address these threats are very limited, resulting in the need for more advanced techniques. Inspired by the defense in depth principle, we propose AnoMili, a novel protection system for the MIL-STD-1553 bus, which consists of: (i) a physical intrusion detection mechanism that detects unauthorized devices connected to the 1553 bus, even if they are passive (sniffing), (ii) a device fingerprinting mechanism that protects against spoofing attacks (two approaches are proposed: prevention and detection), (iii) a context-based anomaly detection mechanism, and (iv) an anomaly explanation engine responsible for explaining the detected anomalies in real time. We evaluate AnoMili's effectiveness and practicability in two real 1553 hardware-based testbeds. The effectiveness of the anomaly explanation engine is also demonstrated. All of the detection and prevention mechanisms employed had high detection rates (over 99.45%) with low false positive rates. The context-based anomaly detection mechanism obtained perfect results when evaluated on a dataset used in prior work.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Evaluating the Security of Open Radio Access Networks

The Open Radio Access Network (O-RAN) is a promising RAN architecture, aimed at reshaping the RAN industry toward an open, adaptive, and intelligent RAN. In this paper, we conducted a comprehensive security analysis of Open Radio Access Networks (O-RAN). Specifically, we review the architectural blueprint designed by the O-RAN alliance -- A leading force in the cellular ecosystem. Within the security analysis, we provide a detailed overview of the O-RAN architecture; present an ontology for evaluating the security of a system, which is currently at an early development stage; detect the primary risk areas to O-RAN; enumerate the various threat actors to O-RAN; and model potential threats to O-RAN. The significance of this work is providing an updated attack surface to cellular network operators. Based on the attack surface, cellular network operators can carefully deploy the appropriate countermeasure for increasing the security of O-RAN.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

EyeDAS: Securing Perception of Autonomous Cars Against the Stereoblindness Syndrome

The ability to detect whether an object is a 2D or 3D object is extremely important in autonomous driving, since a detection error can have life-threatening consequences, endangering the safety of the driver, passengers, pedestrians, and others on the road. Methods proposed to distinguish between 2 and 3D objects (e.g., liveness detection methods) are not suitable for autonomous driving, because they are object dependent or do not consider the constraints associated with autonomous driving (e.g., the need for real-time decision-making while the vehicle is moving). In this paper, we present EyeDAS, a novel few-shot learning-based method aimed at securing an object detector (OD) against the threat posed by the stereoblindness syndrome (i.e., the inability to distinguish between 2D and 3D objects). We evaluate EyeDAS's real-time performance using 2,000 objects extracted from seven YouTube video recordings of street views taken by a dash cam from the driver's seat perspective. When applying EyeDAS to seven state-of-the-art ODs as a countermeasure, EyeDAS was able to reduce the 2D misclassification rate from 71.42-100% to 2.4% with a 3D misclassification rate of 0% (TPR of 1.0). We also show that EyeDAS outperforms the baseline method and achieves an AUC of over 0.999 and a TPR of 1.0 with an FPR of 0.024.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Large-Scale Shill Bidder Detection in E-commerce

User feedback is one of the most effective methods to build and maintain trust in electronic commerce platforms. Unfortunately, dishonest sellers often bend over backward to manipulate users' feedback or place phony bids in order to increase their own sales and harm competitors. The black market of user feedback, supported by a plethora of shill bidders, prospers on top of legitimate electronic commerce. In this paper, we investigate the ecosystem of shill bidders based on large-scale data by analyzing hundreds of millions of users who performed billions of transactions, and we propose a machine-learning-based method for identifying communities of users that methodically provide dishonest feedback. Our results show that (1) shill bidders can be identified with high precision based on their transaction and feedback statistics; and (2) in contrast to legitimate buyers and sellers, shill bidders form cliques to support each other.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

The Security of Deep Learning Defences for Medical Imaging

Deep learning has shown great promise in the domain of medical image analysis. Medical professionals and healthcare providers have been adopting the technology to speed up and enhance their work. These systems use deep neural networks (DNN) which are vulnerable to adversarial samples; images with imperceivable changes that can alter the model's prediction. Researchers have proposed defences which either make a DNN more robust or detect the adversarial samples before they do harm. However, none of these works consider an informed attacker which can adapt to the defence mechanism. We show that an informed attacker can evade five of the current state of the art defences while successfully fooling the victim's deep learning model, rendering these defences useless. We then suggest better alternatives for securing healthcare DNNs from such attacks: (1) harden the system's security and (2) use digital signatures.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Toward Scalable and Unified Example-based Explanation and Outlier Detection

When neural networks are employed for high-stakes decision-making, it is desirable that they provide explanations for their prediction in order for us to understand the features that have contributed to the decision. At the same time, it is important to flag potential outliers for in-depth verification by domain experts. In this work we propose to unify two differing aspects of explainability with outlier detection. We argue for a broader adoption of prototype-based student networks capable of providing an example-based explanation for their prediction and at the same time identify regions of similarity between the predicted sample and the examples. The examples are real prototypical cases sampled from the training set via our novel iterative prototype replacement algorithm. Furthermore, we propose to use the prototype similarity scores for identifying outliers. We compare performances in terms of the classification, explanation quality, and outlier detection of our proposed network with other baselines. We show that our prototype-based networks beyond similarity kernels deliver meaningful explanations and promising outlier detection results without compromising classification accuracy.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

VISAS -- Detecting GPS spoofing attacks against drones by analyzing camera's video stream

In this study, we propose an innovative method for the real-time detection of GPS spoofing attacks targeting drones, based on the video stream captured by a drone's camera. The proposed method collects frames from the video stream and their location (GPS); by calculating the correlation between each frame, our method can identify an attack on a drone. We first analyze the performance of the suggested method in a controlled environment by conducting experiments on a flight simulator that we developed. Then, we analyze its performance in the real world using a DJI drone. Our method can provide different levels of security against GPS spoofing attacks, depending on the detection interval required; for example, it can provide a high level of security to a drone flying at an altitude of 50-100 meters over an urban area at an average speed of 4 km/h in conditions of low ambient light; in this scenario, the method can provide a level of security that detects any GPS spoofing attack in which the spoofed location is a distance of 1-4 meters (an average of 2.5 meters) from the real location.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

FOOD: Fast Out-Of-Distribution Detector

Deep neural networks (DNNs) perform well at classifying inputs associated with the classes they have been trained on, which are known as in distribution inputs. However, out-of-distribution (OOD) inputs pose a great challenge to DNNs and consequently represent a major risk when DNNs are implemented in safety-critical systems. Extensive research has been performed in the domain of OOD detection. However, current state-of-the-art methods for OOD detection suffer from at least one of the following limitations: (1) increased inference time - this limits existing methods' applicability to many real-world applications, and (2) the need for OOD training data - such data can be difficult to acquire and may not be representative enough, thus limiting the ability of the OOD detector to generalize. In this paper, we propose FOOD -- Fast Out-Of-Distribution detector -- an extended DNN classifier capable of efficiently detecting OOD samples with minimal inference time overhead. Our architecture features a DNN with a final Gaussian layer combined with the log likelihood ratio statistical test and an additional output neuron for OOD detection. Instead of using real OOD data, we use a novel method to craft artificial OOD samples from in-distribution data, which are used to train our OOD detector neuron. We evaluate FOOD's detection performance on the SVHN, CIFAR-10, and CIFAR-100 datasets. Our results demonstrate that in addition to achieving state-of-the-art performance, FOOD is fast and applicable to real-world applications.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A New Methodology for Information Security Risk Assessment for Medical Devices and Its Evaluation

As technology advances towards more connected and digital environments, medical devices are becoming increasingly connected to hospital networks and to the Internet, which exposes them, and thus the patients using them, to new cybersecurity threats. Currently, there is a lack of a methodology dedicated to information security risk assessment for medical devices. In this study, we present the Threat identification, ontology-based Likelihood, severity Decomposition, and Risk integration (TLDR) methodology for information security risk assessment for medical devices. The TLDR methodology uses the following steps: (1) identifying the potentially vulnerable components of medical devices, in this case, four different medical imaging devices (MIDs); (2) identifying the potential attacks, in this case, 23 potential attacks on MIDs; (3) mapping the discovered attacks into a known attack ontology - in this case, the Common Attack Pattern Enumeration and Classifications (CAPECs); (4) estimating the likelihood of the mapped CAPECs in the medical domain with the assistance of a panel of senior healthcare Information Security Experts (ISEs); (5) computing the CAPEC-based likelihood estimates of each attack; (6) decomposing each attack into several severity aspects and assigning them weights; (7) assessing the magnitude of the impact of each of the severity aspects for each attack with the assistance of a panel of senior Medical Experts (MEs); (8) computing the composite severity assessments for each attack; and finally, (9) integrating the likelihood and severity of each attack into its risk, and thus prioritizing it. The details of steps six to eight are beyond the scope of the current study; in the current study, we had replaced them by a single step that included asking the panel of MEs [in this case, radiologists], to assess the overall severity for each attack and use it as its severity...

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

An Automated, End-to-End Framework for Modeling Attacks From Vulnerability Descriptions

Attack graphs are one of the main techniques used to automate the risk assessment process. In order to derive a relevant attack graph, up-to-date information on known attack techniques should be represented as interaction rules. Designing and creating new interaction rules is not a trivial task and currently performed manually by security experts. However, since the number of new security vulnerabilities and attack techniques continuously and rapidly grows, there is a need to frequently update the rule set of attack graph tools with new attack techniques to ensure that the set of interaction rules is always up-to-date. We present a novel, end-to-end, automated framework for modeling new attack techniques from textual description of a security vulnerability. Given a description of a security vulnerability, the proposed framework first extracts the relevant attack entities required to model the attack, completes missing information on the vulnerability, and derives a new interaction rule that models the attack; this new rule is integrated within MulVAL attack graph tool. The proposed framework implements a novel pipeline that includes a dedicated cybersecurity linguistic model trained on the the NVD repository, a recurrent neural network model used for attack entity extraction, a logistic regression model used for completing the missing information, and a novel machine learning-based approach for automatically modeling the attacks as MulVAL's interaction rule. We evaluated the performance of each of the individual algorithms, as well as the complete framework and demonstrated its effectiveness.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

ATHAFI: Agile Threat Hunting And Forensic Investigation

Attackers rapidly change their attacks to evade detection. Even the most sophisticated Intrusion Detection Systems that are based on artificial intelligence and advanced data analytic cannot keep pace with the rapid development of new attacks. When standard detection mechanisms fail or do not provide sufficient forensic information to investigate and mitigate attacks, targeted threat hunting performed by competent personnel is used. Unfortunately, many organization do not have enough security analysts to perform threat hunting tasks and today the level of automation of threat hunting is low. In this paper we describe a framework for agile threat hunting and forensic investigation (ATHAFI), which automates the threat hunting process at multiple levels. Adaptive targeted data collection, attack hypotheses generation, hypotheses testing, and continuous threat intelligence feeds allow to perform simple investigations in a fully automated manner. The increased level of automation will significantly boost the analyst's productivity during investigation of the harshest cases. Special Workflow Generation module adapts the threat hunting procedures either to the latest Threat Intelligence obtained from external sources (e.g. National CERT) or to the likeliest attack hypotheses generated by the Attack Hypotheses Generation module. The combination of Attack Hypotheses Generation and Workflows Generation enables intelligent adjustment of workflows, which react to emerging threats effectively.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Autosploit: A Fully Automated Framework for Evaluating the Exploitability of Security Vulnerabilities

The existence of a security vulnerability in a system does not necessarily mean that it can be exploited. In this research, we introduce Autosploit -- an automated framework for evaluating the exploitability of vulnerabilities. Given a vulnerable environment and relevant exploits, Autosploit will automatically test the exploits on different configurations of the environment in order to identify the specific properties necessary for successful exploitation of the existing vulnerabilities. Since testing all possible system configurations is infeasible, we introduce an efficient approach for testing and searching through all possible configurations of the environment. The efficient testing process implemented by Autosploit is based on two algorithms: generalized binary splitting and Barinel, which are used for noiseless and noisy environments respectively. We implemented the proposed framework and evaluated it using real vulnerabilities. The results show that Autosploit is able to automatically identify the system properties that affect the ability to exploit a vulnerability in both noiseless and noisy environments. These important results can be utilized for more accurate and effective risk assessment.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

BRIGHTNESS: Leaking Sensitive Data from Air-Gapped Workstations via Screen Brightness

Air-gapped computers are systems that are kept isolated from the Internet since they store or process sensitive information. In this paper, we introduce an optical covert channel in which an attacker can leak (or, exfiltlrate) sensitive information from air-gapped computers through manipulations on the screen brightness. This covert channel is invisible and it works even while the user is working on the computer. Malware on a compromised computer can obtain sensitive data (e.g., files, images, encryption keys and passwords), and modulate it within the screen brightness, invisible to users. The small changes in the brightness are invisible to humans but can be recovered from video streams taken by cameras such as a local security camera, smartphone camera or a webcam. We present related work and discuss the technical and scientific background of this covert channel. We examined the channel's boundaries under various parameters, with different types of computer and TV screens, and at several distances. We also tested different types of camera receivers to demonstrate the covert channel. Lastly, we present relevant countermeasures to this type of attack. Lastly, we present relevant countermeasures to this type of attack.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

DANTE: A framework for mining and monitoring darknet traffic

Trillions of network packets are sent over the Internet to destinations which do not exist. This 'darknet' traffic captures the activity of botnets and other malicious campaigns aiming to discover and compromise devices around the world. In order to mine threat intelligence from this data, one must be able to handle large streams of logs and represent the traffic patterns in a meaningful way. However, by observing how network ports (services) are used, it is possible to capture the intent of each transmission. In this paper, we present DANTE: a framework and algorithm for mining darknet traffic. DANTE learns the meaning of targeted network ports by applying Word2Vec to observed port sequences. Then, when a host sends a new sequence, DANTE represents the transmission as the average embedding of the ports found that sequence. Finally, DANTE uses a novel and incremental time-series cluster tracking algorithm on observed sequences to detect recurring behaviors and new emerging threats. To evaluate the system, we ran DANTE on a full year of darknet traffic (over three Tera-Bytes) collected by the largest telecommunications provider in Europe, Deutsche Telekom and analyzed the results. DANTE discovered 1,177 new emerging threats and was able to track malicious campaigns over time. We also compared DANTE to the current best approach and found DANTE to be more practical and effective at detecting darknet traffic patterns.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Exploring the Back Alleys: Analysing The Robustness of Alternative Neural Network Architectures against Adversarial Attacks

We investigate to what extent alternative variants of Artificial Neural Networks (ANNs) are susceptible to adversarial attacks. We analyse the adversarial robustness of conventional, stochastic ANNs and Spiking Neural Networks (SNNs) in the raw image space, across three different datasets. Our experiments reveal that stochastic ANN variants are almost equally as susceptible as conventional ANNs when faced with simple iterative gradient-based attacks in the white-box setting. However we observe, that in black-box settings, stochastic ANNs are more robust than conventional ANNs, when faced with boundary attacks, transferability and surrogate attacks. Consequently, we propose improved attacks and defence mechanisms for stochastic ANNs in black-box settings. When performing surrogate-based black-box attacks, one can employ stochastic models as surrogates to observe higher attack success on both stochastic and deterministic targets. This success can be further improved with our proposed Variance Mimicking (VM) surrogate training method, against stochastic targets. Finally, adopting a defender's perspective, we investigate the plausibility of employing stochastic switching of model mixtures as a viable hardening mechanism. We observe that such a scheme does provide a partial hardening.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

GIM: Gaussian Isolation Machines

In many cases, neural network classifiers are likely to be exposed to input data that is outside of their training distribution data. Samples from outside the distribution may be classified as an existing class with high probability by softmax-based classifiers; such incorrect classifications affect the performance of the classifiers and the applications/systems that depend on them. Previous research aimed at distinguishing training distribution data from out-of-distribution data (OOD) has proposed detectors that are external to the classification method. We present Gaussian isolation machine (GIM), a novel hybrid (generative-discriminative) classifier aimed at solving the problem arising when OOD data is encountered. The GIM is based on a neural network and utilizes a new loss function that imposes a distribution on each of the trained classes in the neural network's output space, which can be approximated by a Gaussian. The proposed GIM's novelty lies in its discriminative performance and generative capabilities, a combination of characteristics not usually seen in a single classifier. The GIM achieves state-of-the-art classification results on image recognition and sentiment analysis benchmarking datasets and can also deal with OOD inputs.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Lightweight Collaborative Anomaly Detection for the IoT using Blockchain

Due to their rapid growth and deployment, the Internet of things (IoT) have become a central aspect of our daily lives. Unfortunately, IoT devices tend to have many vulnerabilities which can be exploited by an attacker. Unsupervised techniques, such as anomaly detection, can be used to secure these devices in a plug-and-protect manner. However, anomaly detection models must be trained for a long time in order to capture all benign behaviors. Furthermore, the anomaly detection model is vulnerable to adversarial attacks since, during the training phase, all observations are assumed to be benign. In this paper, we propose (1) a novel approach for anomaly detection and (2) a lightweight framework that utilizes the blockchain to ensemble an anomaly detection model in a distributed environment. Blockchain framework incrementally updates a trusted anomaly detection model via self-attestation and consensus among the IoT devices. We evaluate our method on a distributed IoT simulation platform, which consists of 48 Raspberry Pis. The simulation demonstrates how the approach can enhance the security of each device and the security of the network as a whole.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

9-1-1 DDoS: Threat, Analysis and Mitigation

The 911 emergency service belongs to one of the 16 critical infrastructure sectors in the United States. Distributed denial of service (DDoS) attacks launched from a mobile phone botnet pose a significant threat to the availability of this vital service. In this paper we show how attackers can exploit the cellular network protocols in order to launch an anonymized DDoS attack on 911. The current FCC regulations require that all emergency calls be immediately routed regardless of the caller's identifiers (e.g., IMSI and IMEI). A rootkit placed within the baseband firmware of a mobile phone can mask and randomize all cellular identifiers, causing the device to have no genuine identification within the cellular network. Such anonymized phones can issue repeated emergency calls that cannot be blocked by the network or the emergency call centers, technically or legally. We explore the 911 infrastructure and discuss why it is susceptible to this kind of attack. We then implement different forms of the attack and test our implementation on a small cellular network. Finally, we simulate and analyze anonymous attacks on a model of current 911 infrastructure in order to measure the severity of their impact. We found that with less than 6K bots (or $100K hardware), attackers can block emergency services in an entire state (e.g., North Carolina) for days. We believe that this paper will assist the respective organizations, lawmakers, and security professionals in understanding the scope of this issue in order to prevent possible 911-DDoS attacks in the future.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

DiskFiltration: Data Exfiltration from Speakerless Air-Gapped Computers via Covert Hard Drive Noise

Air-gapped computers are disconnected from the Internet physically and logically. This measure is taken in order to prevent the leakage of sensitive data from secured networks. In the past, it has been shown that malware can exfiltrate data from air-gapped computers by transmitting ultrasonic signals via the computer's speakers. However, such acoustic communication relies on the availability of speakers on a computer. In this paper, we present 'DiskFiltration,' a covert channel which facilitates the leakage of data from an air-gapped compute via acoustic signals emitted from its hard disk drive (HDD). Our method is unique in that, unlike other acoustic covert channels, it doesn't require the presence of speakers or audio hardware in the air-gapped computer. A malware installed on a compromised machine can generate acoustic emissions at specific audio frequencies by controlling the movements of the HDD's actuator arm. Digital Information can be modulated over the acoustic signals and then be picked up by a nearby receiver (e.g., smartphone, smartwatch, laptop, etc.). We examine the HDD anatomy and analyze its acoustical characteristics. We also present signal generation and detection, and data modulation and demodulation algorithms. Based on our proposed method, we developed a transmitter on a personal computer and a receiver on a smartphone, and we provide the design and implementation details. We also evaluate our covert channel on various types of internal and external HDDs in different computer chassis and at various distances. With DiskFiltration we were able to covertly transmit data (e.g., passwords, encryption keys, and keylogging data) between air-gapped computers to a smartphone at an effective bit rate of 180 bits/minute (10,800 bits/hour) and a distance of up to two meters (six feet).

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

dr0wned - Cyber-Physical Attack with Additive Manufacturing

Additive manufacturing (AM), or 3D printing, is an emerging manufacturing technology that is expected to have far-reaching socioeconomic, environmental, and geopolitical implications. As use of this technology increases, it will become more common to produce functional parts, including components for safety-critical systems. AM's dependence on computerization raises the concern that the manufactured part's quality can be compromised by sabotage. This paper demonstrates the validity of this concern, as we present the very first full chain of attack involving AM, beginning with a cyber attack aimed at compromising a benign AM component, continuing with malicious modification of a manufactured object's blueprint, leading to the sabotage of the manufactured functional part, and resulting in the physical destruction of a cyber-physical system that employs this part. The contributions of this paper are as follows. We propose a systematic approach to identify opportunities for an attack involving AM that enables an adversary to achieve his/her goals. Then we propose a methodology to assess the level of difficulty of an attack, thus enabling differentiation between possible attack chains. Finally, to demonstrate the experimental proof for the entire attack chain, we sabotage the 3D printed propeller of a quadcopter UAV, causing the quadcopter to literally fall from the sky.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Fansmitter: Acoustic Data Exfiltration from (Speakerless) Air-Gapped Computers

Because computers may contain or interact with sensitive information, they are often air-gapped and in this way kept isolated and disconnected from the Internet. In recent years the ability of malware to communicate over an air-gap by transmitting sonic and ultrasonic signals from a computer speaker to a nearby receiver has been shown. In order to eliminate such acoustic channels, current best practice recommends the elimination of speakers (internal or external) in secure computers, thereby creating a so-called 'audio-gap'. In this paper, we present Fansmitter, a malware that can acoustically exfiltrate data from air-gapped computers, even when audio hardware and speakers are not present. Our method utilizes the noise emitted from the CPU and chassis fans which are present in virtually every computer today. We show that a software can regulate the internal fans' speed in order to control the acoustic waveform emitted from a computer. Binary data can be modulated and transmitted over these audio signals to a remote microphone (e.g., on a nearby mobile phone). We present Fansmitter's design considerations, including acoustic signature analysis, data modulation, and data transmission. We also evaluate the acoustic channel, present our results, and discuss countermeasures. Using our method we successfully transmitted data from air-gapped computer without audio hardware, to a smartphone receiver in the same room. We demonstrated the effective transmission of encryption keys and passwords from a distance of zero to eight meters, with bit rate of up to 900 bits/hour. We show that our method can also be used to leak data from different types of IT equipment, embedded systems, and IoT devices that have no audio hardware, but contain fans of various types and sizes.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Handwritten Signature Verification Using Hand-Worn Devices

Online signature verification technologies, such as those available in banks and post offices, rely on dedicated digital devices such as tablets or smart pens to capture, analyze and verify signatures. In this paper, we suggest a novel method for online signature verification that relies on the increasingly available hand-worn devices, such as smartwatches or fitness trackers, instead of dedicated ad-hoc devices. Our method uses a set of known genuine and forged signatures, recorded using the motion sensors of a hand-worn device, to train a machine learning classifier. Then, given the recording of an unknown signature and a claimed identity, the classifier can determine whether the signature is genuine or forged. In order to validate our method, it was applied on 1980 recordings of genuine and forged signatures that we collected from 66 subjects in our institution. Using our method, we were able to successfully distinguish between genuine and forged signatures with a high degree of accuracy (0.98 AUC and 0.05 EER).

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Personal Information Leakage During Password Recovery of Internet Services

In this paper we examine the standard password recovery process of large Internet services such as Gmail, Facebook, and Twitter. Although most of these services try to maintain user privacy, with regard to registration information and other personal information provided by the user, we demonstrate that personal information can still be obtained by unauthorized individuals or attackers. This information includes the full (or partial) email address, phone number, friends list, address, etc. We examine different scenarios and demonstrate how the details revealed in the password recovery process can be used to deduct more focused information about users.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Security Testbed for the Internet of Things

The Internet of Things (IoT) is a global ecosystem of information and communication technologies aimed at connecting any type of object (thing), at any time and in any place, to each other and to the Internet. One of the major problems associated with the IoT is maintaining security; the heterogeneous nature of such deployments poses a challenge to many aspects of security, including security testing and analysis. In addition, there is no existing mechanism that performs security testing for IoT devices in different contexts. In this paper, we propose an innovative security testbed framework targeted at IoT devices. The security testbed supports both standard and context-based security testing, with a set of security tests conducted under the different environmental conditions in which IoT devices operate. The requirements and architectural design of the proposed testbed are discussed, and the testbed operation is demonstrated in several testing scenarios.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

SPEAKE(a)R: Turn Speakers to Microphones for Fun and Profit

It is possible to manipulate the headphones (or earphones) connected to a computer, silently turning them into a pair of eavesdropping microphones - with software alone. The same is also true for some types of loudspeakers. This paper focuses on this threat in a cyber-security context. We present SPEAKE(a)R, a software that can covertly turn the headphones connected to a PC into a microphone. We present technical background and explain why most of PCs and laptops are susceptible to this type of attack. We examine an attack scenario in which malware can use a computer as an eavesdropping device, even when a microphone is not present, muted, taped, or turned off. We measure the signal quality and the effective distance, and survey the defensive countermeasures.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

USBee: Air-Gap Covert-Channel via Electromagnetic Emission from USB

In recent years researchers have demonstrated how attackers could use USB connectors implanted with RF transmitters to exfiltrate data from secure, and even air-gapped, computers (e.g., COTTONMOUTH in the leaked NSA ANT catalog). Such methods require a hardware modification of the USB plug or device, in which a dedicated RF transmitter is embedded. In this paper we present USBee, a software that can utilize an unmodified USB device connected to a computer as a RF transmitter. We demonstrate how a software can intentionally generate controlled electromagnetic emissions from the data bus of a USB connector. We also show that the emitted RF signals can be controlled and modulated with arbitrary binary data. We implement a prototype of USBee, and discuss its design and implementation details including signal generation and modulation. We evaluate the transmitter by building a receiver and demodulator using GNU Radio. Our evaluation shows that USBee can be used for transmitting binary data to a nearby receiver at a bandwidth of 20 to 80 BPS (bytes per second).

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Virtual Breathalyzer

Driving under the influence of alcohol is a widespread phenomenon in the US where it is considered a major cause of fatal accidents. In this research we present a novel approach and concept for detecting intoxication from motion differences obtained by the sensors of wearable devices. We formalize the problem of drunkenness detection as a supervised machine learning task, both as a binary classification problem (drunk or sober) and a regression problem (the breath alcohol content level). In order to test our approach, we collected data from 30 different subjects (patrons at three bars) using Google Glass and the LG G-watch, Microsoft Band, and Samsung Galaxy S4. We validated our results against an admissible breathalyzer used by the police. A system based on this concept, successfully detected intoxication and achieved the following results: 0.95 AUC and 0.05 FPR, given a fixed TPR of 1.0. Applications based on our system can be used to analyze the free gait of drinkers when they walk from the car to the bar and vice-versa, in order to alert people, or even a connected car and prevent people from driving under the influence of alcohol.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

VisiSploit: An Optical Covert-Channel to Leak Data through an Air-Gap

In recent years, various out-of-band covert channels have been proposed that demonstrate the feasibility of leaking data out of computers without the need for network connectivity. The methods proposed have been based on different type of electromagnetic, acoustic, and thermal emissions. However, optical channels have largely been considered less covert: because they are visible to the human eye and hence can be detected, they have received less attention from researchers. In this paper, we introduce VisiSploit, a new type of optical covert channel which, unlike other optical methods, is also stealthy. Our method exploits the limitations of human visual perception in order to unobtrusively leak data through a standard computer LCD display. Our experiments show that very low contrast or fast flickering images which are invisible to human subjects, can be recovered from photos taken by a camera. Consequentially, we show that malicious code on a compromised computer can obtain sensitive data (e.g., images, encryption keys, passwords), and project it onto a computer LCD screen, invisible and unbeknownst to users, allowing an attacker to reconstruct the data using a photo taken by a nearby (possibly hidden) camera. In order to demonstrate the feasibility of this type of attack and evaluate the channel's stealth, we conducted a battery of tests with 40 human subjects. We also examined the channel's boundaries under various parameters, with different types of encoded objects, at several distances, and using several kinds of cameras. Our results show that binary data can be leaked via our covert channel. Further research and discussion may widen the scope of this field beyond its current boundaries, yielding novel attack paradigms that exploit the subtle mechanisms of human visual perception.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

BitWhisper: Covert Signaling Channel between Air-Gapped Computers using Thermal Manipulations

It has been assumed that the physical separation (air-gap) of computers provides a reliable level of security, such that should two adjacent computers become compromised, the covert exchange of data between them would be impossible. In this paper, we demonstrate BitWhisper, a method of bridging the air-gap between adjacent compromised computers by using their heat emissions and built-in thermal sensors to create a covert communication channel. Our method is unique in two respects: it supports bidirectional communication, and it requires no additional dedicated peripheral hardware. We provide experimental results based on implementation of BitWhisper prototype, and examine the channel properties and limitations. Our experiments included different layouts, with computers positioned at varying distances from one another, and several sensor types and CPU configurations (e.g., Virtual Machines). We also discuss signal modulation and communication protocols, showing how BitWhisper can be used for the exchange of data between two computers in a close proximity (at distance of 0-40cm) at an effective rate of 1-8 bits per hour, a rate which makes it possible to infiltrate brief commands and exfiltrate small amount of data (e.g., passwords) over the covert channel.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

JoKER: Trusted Detection of Kernel Rootkits in Android Devices via JTAG Interface

Smartphones and tablets have become prime targets for malware, due to the valuable private and corporate information they hold. While Anti-Virus (AV) program may successfully detect malicious applications (apps), they remain ineffective against low-level rootkits that evade detection mechanisms by masking their own presence. Furthermore, any detection mechanism run on the same physical device as the monitored OS can be compromised via application, kernel or boot-loader vulnerabilities. Consequentially, trusted detection of kernel rootkits in mobile devices is a challenging task in practice. In this paper we present JoKER - a system which aims at detecting rootkits in the Android kernel by utilizing the hardware's Joint Test Action Group (JTAG) interface for trusted memory forensics. Our framework consists of components that extract areas of a kernel's memory and reconstruct it for further analysis. We present the overall architecture along with its implementation, and demonstrate that the system can successfully detect the presence of stealthy rootkits in the kernel. The results show that although JTAG's main purpose is system testing, it can also be used for malware detection where traditional methods fail.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

AirHopper: Bridging the Air-Gap between Isolated Networks and Mobile Phones using Radio Frequencies

Information is the most critical asset of modern organizations, and accordingly coveted by adversaries. When highly sensitive data is involved, an organization may resort to air-gap isolation, in which there is no networking connection between the inner network and the external world. While infiltrating an air-gapped network has been proven feasible in recent years (e.g., Stuxnet), data exfiltration from an air-gapped network is still considered to be one of the most challenging phases of an advanced cyber-attack. In this paper we present "AirHopper", a bifurcated malware that bridges the air-gap between an isolated network and nearby infected mobile phones using FM signals. While it is known that software can intentionally create radio emissions from a video display unit, this is the first time that mobile phones are considered in an attack model as the intended receivers of maliciously crafted radio signals. We examine the attack model and its limitations, and discuss implementation considerations such as stealth and modulation methods. Finally, we evaluate AirHopper and demonstrate how textual and binary data can be exfiltrated from physically isolated computer to mobile phones at a distance of 1-7 meters, with effective bandwidth of 13-60 Bps (Bytes per second).

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Data Mining of Online Genealogy Datasets for Revealing Lifespan Patterns in Human Population

Online genealogy datasets contain extensive information about millions of people and their past and present family connections. This vast amount of data can assist in identifying various patterns in human population. In this study, we present methods and algorithms which can assist in identifying variations in lifespan distributions of human population in the past centuries, in detecting social and genetic features which correlate with human lifespan, and in constructing predictive models of human lifespan based on various features which can easily be extracted from genealogy datasets. We have evaluated the presented methods and algorithms on a large online genealogy dataset with over a million profiles and over 9 million connections, all of which were collected from the WikiTree website. Our findings indicate that significant but small positive correlations exist between the parents' lifespan and their children's lifespan. Additionally, we found slightly higher and significant correlations between the lifespans of spouses. We also discovered a very small positive and significant correlation between longevity and reproductive success in males, and a small and significant negative correlation between longevity and reproductive success in females. Moreover, our machine learning algorithms presented better than random classification results in predicting which people who outlive the age of 50 will also outlive the age of 80. We believe that this study will be the first of many studies which utilize the wealth of data on human populations, existing in online genealogy datasets, to better understand factors which influence human lifespan. Understanding these factors can assist scientists in providing solutions for successful aging.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Matching Entities Across Online Social Networks

Online Social Networks (OSNs), such as Facebook and Twitter, have become an integral part of our daily lives. There are hundreds of OSNs, each with its own focus in that each offers particular services and functionalities. Recent studies show that many OSN users create several accounts on multiple OSNs using the same or different personal information. Collecting all the available data of an individual from several OSNs and fusing it into a single profile can be useful for many purposes. In this paper, we introduce novel machine learning based methods for solving Entity Resolution (ER), a problem for matching user profiles across multiple OSNs. The presented methods are able to match between two user profiles from two different OSNs based on supervised learning techniques, which use features extracted from each one of the user profiles. By using the extracted features and supervised learning techniques, we developed classifiers which can perform entity matching between two profiles for the following scenarios: (a) matching entities across two OSNs; (b) searching for a user by similar name; and (c) de-anonymizing a user's identity. The constructed classifiers were tested by using data collected from two popular OSNs, Facebook and Xing. We then evaluated the classifiers' performances using various evaluation measures, such as true and false positive rates, accuracy, and the Area Under the receiver operator Curve (AUC). The constructed classifiers were evaluated and their classification performance measured by AUC was quite remarkable, with an AUC of up to 0.982 and an accuracy of up to 95.9% in identifying user profiles across two OSNs.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Online Social Networks: Threats and Solutions

Many online social network (OSN) users are unaware of the numerous security risks that exist in these networks, including privacy violations, identity theft, and sexual harassment, just to name a few. According to recent studies, OSN users readily expose personal and private details about themselves, such as relationship status, date of birth, school name, email address, phone number, and even home address. This information, if put into the wrong hands, can be used to harm users both in the virtual world and in the real world. These risks become even more severe when the users are children. In this paper we present a thorough review of the different security and privacy risks which threaten the well-being of OSN users in general, and children in particular. In addition, we present an overview of existing solutions that can provide better protection, security, and privacy for OSN users. We also offer simple-to-implement recommendations for OSN users which can improve their security and privacy when using these platforms. Furthermore, we suggest future research directions.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Quantitative Analysis of Genealogy Using Digitised Family Trees

Driven by the popularity of television shows such as Who Do You Think You Are? many millions of users have uploaded their family tree to web projects such as WikiTree. Analysis of this corpus enables us to investigate genealogy computationally. The study of heritage in the social sciences has led to an increased understanding of ancestry and descent but such efforts are hampered by difficult to access data. Genealogical research is typically a tedious process involving trawling through sources such as birth and death certificates, wills, letters and land deeds. Decades of research have developed and examined hypotheses on population sex ratios, marriage trends, fertility, lifespan, and the frequency of twins and triplets. These can now be tested on vast datasets containing many billions of entries using machine learning tools. Here we survey the use of genealogy data mining using family trees dating back centuries and featuring profiles on nearly 7 million individuals based in over 160 countries. These data are not typically created by trained genealogists and so we verify them with reference to third party censuses. We present results on a range of aspects of population dynamics. Our approach extends the boundaries of genealogy inquiry to precise measurement of underlying human phenomena.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Reaction to New Security Threat Class

Each new identified security threat class triggers new research and development efforts by the scientific and professional communities. In this study, we investigate the rate at which the scientific and professional communities react to new identified threat classes as it is reflected in the number of patents, scientific articles and professional publications over a long period of time. The following threat classes were studied: Phishing; SQL Injection; BotNet; Distributed Denial of Service; and Advanced Persistent Threat. Our findings suggest that in most cases it takes a year for the scientific community and more than two years for industry to react to a new threat class with patents. Since new products follow patents, it is reasonable to expect that there will be a window of approximately two to three years in which no effective product is available to cope with the new threat class.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

ACTIDS: An Active Strategy For Detecting And Localizing Network Attacks

In this work we investigate a new approach for detecting attacks which aim to degrade the network's Quality of Service (QoS). To this end, a new network-based intrusion detection system (NIDS) is proposed. Most contemporary NIDSs take a passive approach by solely monitoring the network's production traffic. This paper explores a complementary approach in which distributed agents actively send out periodic probes. The probes are continuously monitored to detect anomalous behavior of the network. The proposed approach takes away much of the variability of the network's production traffic that makes it so difficult to classify. This enables the NIDS to detect more subtle attacks which would not be detected using the passive approach alone. Furthermore, the active probing approach allows the NIDS to be effectively trained using only examples of the network's normal states, hence enabling an effective detection of zero-day attacks. Using realistic experiments, we show that an NIDS which also leverages the active approach is considerably more effective in detecting attacks which aim to degrade the network's QoS when compared to an NIDS which relies solely on the passive approach. Lastly, we show that the false positives rate remains very low even in the face of Byzantine faults.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Combining One-Class Classifiers via Meta-Learning

Selecting the best classifier among the available ones is a difficult task, especially when only instances of one class exist. In this work we examine the notion of combining one-class classifiers as an alternative for selecting the best classifier. In particular, we propose two new one-class classification performance measures to weigh classifiers and show that a simple ensemble that implements these measures can outperform the most popular one-class ensembles. Furthermore, we propose a new one-class ensemble scheme, TUPSO, which uses meta-learning to combine one-class classifiers. Our experiments demonstrate the superiority of TUPSO over all other tested ensembles and show that the TUPSO performance is statistically indistinguishable from that of the hypothetical best classifier.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Ethical Considerations when Employing Fake Identities in OSN for Research

Online Social Networks (OSNs) have rapidly become a prominent and widely used service, offering a wealth of personal and sensitive information with significant security and privacy implications. Hence, OSNs are also an important - and popular - subject for research. To perform research based on real-life evidence, however, researchers may need to access OSN data, such as texts and files uploaded by users and connections among users. This raises significant ethical problems. Currently, there are no clear ethical guidelines, and researchers may end up (unintentionally) performing ethically questionable research, sometimes even when more ethical research alternatives exist. For example, several studies have employed `fake identities` to collect data from OSNs, but fake identities may be used for attacks and are considered a security issue. Is it legitimate to use fake identities for studying OSNs or for collecting OSN data for research? We present a taxonomy of the ethical challenges facing researchers of OSNs and compare different approaches. We demonstrate how ethical considerations have been taken into account in previous studies that used fake identities. In addition, several possible approaches are offered to reduce or avoid ethical misconducts. We hope this work will stimulate the development and use of ethical practices and methods in the research of online social networks.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Facebook Applications' Installation and Removal: A Temporal Analysis

Facebook applications are one of the reasons for Facebook attractiveness. Unfortunately, numerous users are not aware of the fact that many malicious Facebook applications exist. To educate users, to raise users' awareness and to improve Facebook users' security and privacy, we developed a Firefox add-on that alerts users to the number of installed applications on their Facebook profiles. In this study, we present the temporal analysis of the Facebook applications' installation and removal dataset collected by our add-on. This dataset consists of information from 2,945 users, collected during a period of over a year. We used linear regression to analyze our dataset and discovered the linear connection between the average percentage change of newly installed Facebook applications and the number of days passed since the user initially installed our add-on. Additionally, we found out that users who used our Firefox add-on become more aware of their security and privacy installing on average fewer new applications. Finally, we discovered that on average 86.4% of Facebook users install an additional application every 4.2 days.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Friend or Foe? Fake Profile Identification in Online Social Networks

The amount of personal information unwillingly exposed by users on online social networks is staggering, as shown in recent research. Moreover, recent reports indicate that these networks are infested with tens of millions of fake users profiles, which may jeopardize the users' security and privacy. To identify fake users in such networks and to improve users' security and privacy, we developed the Social Privacy Protector software for Facebook. This software contains three protection layers, which improve user privacy by implementing different methods. The software first identifies a user's friends who might pose a threat and then restricts this "friend's" exposure to the user's personal information. The second layer is an expansion of Facebook's basic privacy settings based on different types of social network usage profiles. The third layer alerts users about the number of installed applications on their Facebook profile, which have access to their private information. An initial version of the Social Privacy Protection software received high media coverage, and more than 3,000 users from more than twenty countries have installed the software, out of which 527 used the software to restrict more than nine thousand friends. In addition, we estimate that more than a hundred users accepted the software's recommendations and removed at least 1,792 Facebook applications from their profiles. By analyzing the unique dataset obtained by the software in combination with machine learning techniques, we developed classifiers, which are able to predict which Facebook profiles have high probabilities of being fake and therefore, threaten the user's well-being. Moreover, in this study, we present statistics on users' privacy settings and statistics of the number of applications installed on Facebook profiles...

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Organization Mining Using Online Social Networks

Mature social networking services are one of the greatest assets of today's organizations. This valuable asset, however, can also be a threat to an organization's confidentiality. Members of social networking websites expose not only their personal information, but also details about the organizations for which they work. In this paper we analyze several commercial organizations by mining data which their employees have exposed on Facebook, LinkedIn, and other publicly available sources. Using a web crawler designed for this purpose, we extract a network of informal social relationships among employees of a given target organization. Our results, obtained using centrality analysis and Machine Learning techniques applied to the structure of the informal relationships network, show that it is possible to identify leadership roles within the organization solely by this means. It is also possible to gain valuable non-trivial insights on an organization's structure by clustering its social network and gathering publicly available information on the employees within each cluster. Organizations wanting to conceal their internal structure, identity of leaders, location and specialization of branches offices, etc., must enforce strict policies to control the use of social media by their employees.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Securing Your Transactions: Detecting Anomalous Patterns In XML Documents

XML transactions are used in many information systems to store data and interact with other systems. Abnormal transactions, the result of either an on-going cyber attack or the actions of a benign user, can potentially harm the interacting systems and therefore they are regarded as a threat. In this paper we address the problem of anomaly detection and localization in XML transactions using machine learning techniques. We present a new XML anomaly detection framework, XML-AD. Within this framework, an automatic method for extracting features from XML transactions was developed as well as a practical method for transforming XML features into vectors of fixed dimensionality. With these two methods in place, the XML-AD framework makes it possible to utilize general learning algorithms for anomaly detection. Central to the functioning of the framework is a novel multi-univariate anomaly detection algorithm, ADIFA. The framework was evaluated on four XML transactions datasets, captured from real information systems, in which it achieved over 89% true positive detection rate with less than a 0.2% false positive rate.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Incremental Learning with Accuracy Prediction of Social and Individual Properties from Mobile-Phone Data

Mobile phones are quickly becoming the primary source for social, behavioral, and environmental sensing and data collection. Today's smartphones are equipped with increasingly more sensors and accessible data types that enable the collection of literally dozens of signals related to the phone, its user, and its environment. A great deal of research effort in academia and industry is put into mining this raw data for higher level sense-making, such as understanding user context, inferring social networks, learning individual features, predicting outcomes, and so on. In this work we investigate the properties of learning and inference of real world data collected via mobile phones over time. In particular, we look at the dynamic learning process over time, and how the ability to predict individual parameters and social links is incrementally enhanced with the accumulation of additional data. To do this, we use the Friends and Family dataset, which contains rich data signals gathered from the smartphones of 140 adult members of a young-family residential community for over a year, and is one of the most comprehensive mobile phone datasets gathered in academia to date. We develop several models that predict social and individual properties from sensed mobile phone data, including detection of life-partners, ethnicity, and whether a person is a student or not. Then, for this set of diverse learning tasks, we investigate how the prediction accuracy evolves over time, as new data is collected. Finally, based on gained insights, we propose a method for advance prediction of the maximal learning accuracy possible for the learning task at hand, based on an initial set of measurements. This has practical implications, like informing the design of mobile data collection campaigns, or evaluating analysis strategies.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Efficient Collaborative Application Monitoring Scheme for Mobile Networks

New operating systems for mobile devices allow their users to download millions of applications created by various individual programmers, some of which may be malicious or flawed. In order to detect that an application is malicious, monitoring its operation in a real environment for a significant period of time is often required. Mobile devices have limited computation and power resources and thus are limited in their monitoring capabilities. In this paper we propose an efficient collaborative monitoring scheme that harnesses the collective resources of many mobile devices, "vaccinating" them against potentially unsafe applications. We suggest a new local information flooding algorithm called "TTL Probabilistic Propagation" (TPP). The algorithm periodically monitors one or more application and reports its conclusions to a small number of other mobile devices, who then propagate this information onwards. The algorithm is analyzed, and is shown to outperform existing state of the art information propagation algorithms, in terms of convergence time as well as network overhead. The maximal "load" of the algorithm (the fastest arrival rate of new suspicious applications, that can still guarantee complete monitoring), is analytically calculated and shown to be significantly superior compared to any non-collaborative approach. Finally, we show both analytically and experimentally using real world network data that implementing the proposed algorithm significantly reduces the number of infected mobile devices. In addition, we analytically prove that the algorithm is tolerant to several types of Byzantine attacks where some adversarial agents may generate false information, or abuse the algorithm in other ways.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Stealing Reality

In this paper we discuss the threat of malware targeted at extracting information about the relationships in a real-world social network as well as characteristic information about the individuals in the network, which we dub Stealing Reality. We present Stealing Reality, explain why it differs from traditional types of network attacks, and discuss why its impact is significantly more dangerous than that of other attacks. We also present our initial analysis and results regarding the form that an SR attack might take, with the goal of promoting the discussion of defending against such an attack, or even just detecting the fact that one has already occurred.

0 citations0 reviewsNo code linkedOpen access