Biblio

As the era of cloud technology arises, more and more people are beginning to migrate their applications and personal data to the cloud. This makes web-based applications an attractive target for cyber-attacks. As a result, web-based applications now need more protections than ever. However, current anomaly-based web attack detection approaches face the difficulties like unsatisfying accuracy and lack of generalization. And the rule-based web attack detection can hardly fight unknown attacks and is relatively easy to bypass. Therefore, we propose a novel deep learning approach to detect anomalous requests. Our approach is to first train two Recurrent Neural Networks (RNNs) with the complicated recurrent unit (LSTM unit or GRU unit) to learn the normal request patterns using only normal requests unsupervisedly and then supervisedly train a neural network classifier which takes the output of RNNs as the input to discriminate between anomalous and normal requests. We tested our model on two datasets and the results showed that our model was competitive with the state-of-the-art. Our approach frees us from feature selection. Also to the best of our knowledge, this is the first time that the RNN is applied on anomaly-based web attack detection systems.

Data mining and information extraction from data is a field that has gained relevance in recent years thanks to techniques based on artificial intelligence and use of machine and deep learning. The main aim of the present work is the development of a tool based on a previous behaviour study of security audit tools (oriented to SQL pentesting) with the purpose of creating testing sets capable of performing an accurate detection of a SQL attack. The study is based on the information collected through the generated web server logs in a pentesting laboratory environment. Then, making use of the common extracted patterns from the logs, each attack vector has been classified in risk levels (dangerous attack, normal attack, non-attack, etc.). Finally, a training with the generated data was performed in order to obtain a classifier system that has a variable performance between 97 and 99 percent in positive attack detection. The training data is shared to other servers in order to create a distributed network capable of deciding if a query is an attack or is a real petition and inform to connected clients in order to block the petitions from the attacker's IP.

Following the recent adoption of deep neural networks (DNN) accross a wide range of applications, adversarial attacks against these models have proven to be an indisputable threat. Adversarial samples are crafted with a deliberate intention of undermining a system. In the case of DNNs, the lack of better understanding of their working has prevented the development of efficient defenses. In this paper, we propose a new defense method based on practical observations which is easy to integrate into models and performs better than state-of-the-art defenses. Our proposed solution is meant to reinforce the structure of a DNN, making its prediction more stable and less likely to be fooled by adversarial samples. We conduct an extensive experimental study proving the efficiency of our method against multiple attacks, comparing it to numerous defenses, both in white-box and black-box setups. Additionally, the implementation of our method brings almost no overhead to the training procedure, while maintaining the prediction performance of the original model on clean samples.

In this paper, we focus on fraud detection on a signed graph with only a small set of labeled training data. We propose a novel framework that combines deep neural networks and spectral graph analysis. In particular, we use the node projection (called as spectral coordinate) in the low dimensional spectral space of the graph's adjacency matrix as the input of deep neural networks. Spectral coordinates in the spectral space capture the most useful topology information of the network. Due to the small dimension of spectral coordinates (compared with the dimension of the adjacency matrix derived from a graph), training deep neural networks becomes feasible. We develop and evaluate two neural networks, deep autoencoder and convolutional neural network, in our fraud detection framework. Experimental results on a real signed graph show that our spectrum based deep neural networks are effective in fraud detection.

Deep neural networks (DNNs) are one of the most prominent technologies of our time, as they achieve state-of-the-art performance in many machine learning tasks, including but not limited to image classification, text mining, and speech processing. However, recent research on DNNs has indicated ever-increasing concern on the robustness to adversarial examples, especially for security-critical tasks such as traffic sign identification for autonomous driving. Studies have unveiled the vulnerability of a well-trained DNN by demonstrating the ability of generating barely noticeable (to both human and machines) adversarial images that lead to misclassification. Furthermore, researchers have shown that these adversarial images are highly transferable by simply training and attacking a substitute model built upon the target model, known as a black-box attack to DNNs. Similar to the setting of training substitute models, in this paper we propose an effective black-box attack that also only has access to the input (images) and the output (confidence scores) of a targeted DNN. However, different from leveraging attack transferability from substitute models, we propose zeroth order optimization (ZOO) based attacks to directly estimate the gradients of the targeted DNN for generating adversarial examples. We use zeroth order stochastic coordinate descent along with dimension reduction, hierarchical attack and importance sampling techniques to efficiently attack black-box models. By exploiting zeroth order optimization, improved attacks to the targeted DNN can be accomplished, sparing the need for training substitute models and avoiding the loss in attack transferability. Experimental results on MNIST, CIFAR10 and ImageNet show that the proposed ZOO attack is as effective as the state-of-the-art white-box attack (e.g., Carlini and Wagner's attack) and significantly outperforms existing black-box attacks via substitute models.

The Named-Data Networking (NDN) has emerged as a clean-slate Internet proposal on the wave of Information-Centric Networking. Although the NDN's data-plane seems to offer many advantages, e.g., native support for multicast communications and flow balance, it also makes the network infrastructure vulnerable to a specific DDoS attack, the Interest Flooding Attack (IFA). In IFAs, a botnet issuing unsatisfiable content requests can be set up effortlessly to exhaust routers' resources and cause a severe performance drop to legitimate users. So far several countermeasures have addressed this security threat, however, their efficacy was proved by means of simplistic assumptions on the attack model. Therefore, we propose a more complete attack model and design an advanced IFA. We show the efficiency of our novel attack scheme by extensively assessing some of the state-of-the-art countermeasures. Further, we release the software to perform this attack as open source tool to help design future more robust defense mechanisms.

Emerging computing relies heavily on secure backend storage for the massive size of big data originating from the Internet of Things (IoT) smart devices to the Cloud-hosted web applications. Structured Query Language (SQL) Injection Attack (SQLIA) remains an intruder's exploit of choice to pilfer confidential data from the back-end database with damaging ramifications. The existing approaches were all before the new emerging computing in the context of the Internet big data mining and as such will lack the ability to cope with new signatures concealed in a large volume of web requests over time. Also, these existing approaches were strings lookup approaches aimed at on-premise application domain boundary, not applicable to roaming Cloud-hosted services' edge Software-Defined Network (SDN) to application endpoints with large web request hits. Using a Machine Learning (ML) approach provides scalable big data mining for SQLIA detection and prevention. Unfortunately, the absence of corpus to train a classifier is an issue well known in SQLIA research in applying Artificial Intelligence (AI) techniques. This paper presents an application context pattern-driven corpus to train a supervised learning model. The model is trained with ML algorithms of Two-Class Support Vector Machine (TC SVM) and Two-Class Logistic Regression (TC LR) implemented on Microsoft Azure Machine Learning (MAML) studio to mitigate SQLIA. This scheme presented here, then forms the subject of the empirical evaluation in Receiver Operating Characteristic (ROC) curve.

Services today are configured through policies that capture expected behaviors. However, because of subtle and changing stakeholder requirements, producing and maintaining policies is nontrivial. Policy errors are surprisingly common and cause avoidable security vulnerabilities.

We propose Aragorn, an approach that applies formal argumentation to produce policies that balance stakeholder concerns. We demonstrate empirically that, compared to the traditional approach for specifying policies, Aragorn performs (1) better on coverage, correctness, and quality; (2) equally well on learnability and effort÷coverage and difficulty; and (3) slightly worse on time and effort needed. Thus, Aragorn demonstrates the potential for capturing policy rationales as arguments.

We seek to address the challenge of engineering socially intelligent personal agents that are privacy-aware. We propose Arnor, a method, including a metamodel based on social constructs. Arnor incorporates social norms and goes beyond existing agent-oriented software engineering (AOSE) methods by systematically capturing how a personal agent’s actions influence the social experience it delivers. We conduct two empirical studies to evaluate Arnor. First, via a multiphase developer study, we show that Arnor simplifies application development. Second, via simulation experiments, we show that Arnor provides improved privacy-preserving social experience to end users than personal agents engineered using a traditional AOSE method.

Onion sites on the darkweb operate using the Tor Hidden Service (HS) protocol to shield their locations on the Internet, which (among other features) enables these sites to host malicious and illegal content while being resistant to legal action and seizure. Identifying and monitoring such illicit sites in the darkweb is of high relevance to the Computer Security and Law Enforcement communities. We have developed an automated infrastructure that crawls and indexes content from onion sites into a large-scale data repository, called LIGHTS, with over 100M pages. In this paper we describe Automated Tool for Onion Labeling (ATOL), a novel scalable analysis service developed to conduct a thematic assessment of the content of onion sites in the LIGHTS repository. ATOL has three core components – (a) a novel keyword discovery mechanism (ATOLKeyword) which extends analyst-provided keywords for different categories by suggesting new descriptive and discriminative keywords that are relevant for the categories; (b) a classification framework (ATOLClassify) that uses the discovered keywords to map onion site content to a set of categories when sufficient labeled data is available; (c) a clustering framework (ATOLCluster) that can leverage information from multiple external heterogeneous knowledge sources, ranging from domain expertise to Bitcoin transaction data, to categorize onion content in the absence of sufficient supervised data. The paper presents empirical results of ATOL on onion datasets derived from the LIGHTS repository, and additionally benchmarks ATOL's algorithms on the publicly available 20 Newsgroups dataset to demonstrate the reproducibility of its results. On the LIGHTS dataset, ATOLClassify gives a 12% performance gain over an analyst-provided baseline, while ATOLCluster gives a 7% improvement over state-of-the-art semi-supervised clustering algorithms. We also discuss how ATOL has been deployed and externally evaluated, as part of the LIGHTS system.

Testing and fixing Web Application Firewalls (WAFs) are two relevant and complementary challenges for security analysts. Automated testing helps to cost-effectively detect vulnerabilities in a WAF by generating effective test cases, i.e., attacks. Once vulnerabilities have been identified, the WAF needs to be fixed by augmenting its rule set to filter attacks without blocking legitimate requests. However, existing research suggests that rule sets are very difficult to understand and too complex to be manually fixed. In this paper, we formalise the problem of fixing vulnerable WAFs as a combinatorial optimisation problem. To solve it, we propose an automated approach that combines machine learning with multi-objective genetic algorithms. Given a set of legitimate requests and bypassing SQL injection attacks, our approach automatically infers regular expressions that, when added to the WAF's rule set, prevent many attacks while letting legitimate requests go through. Our empirical evaluation based on both open-source and proprietary WAFs shows that the generated filter rules are effective at blocking previously identified and successful SQL injection attacks (recall between 54.6% and 98.3%), while triggering in most cases no or few false positives (false positive rate between 0% and 2%).

The Bonseyes EU H2020 collaborative project aims to develop a platform consisting of a Data Marketplace, a Deep Learning Toolbox, and Developer Reference Platforms for organizations wanting to adopt Artificial Intelligence. The project will be focused on using artificial intelligence in low power Internet of Things (IoT) devices ("edge computing"), embedded computing systems, and data center servers ("cloud computing"). It will bring about orders of magnitude improvements in efficiency, performance, reliability, security, and productivity in the design and programming of systems of artificial intelligence that incorporate Smart Cyber-Physical Systems (CPS). In addition, it will solve a causality problem for organizations who lack access to Data and Models. Its open software architecture will facilitate adoption of the whole concept on a wider scale. To evaluate the effectiveness, technical feasibility, and to quantify the real-world improvements in efficiency, security, performance, effort and cost of adding AI to products and services using the Bonseyes platform, four complementary demonstrators will be built. Bonseyes platform capabilities are aimed at being aligned with the European FI-PPP activities and take advantage of its flagship project FIWARE. This paper provides a description of the project motivation, goals and preliminary work.

An operating system kernel written in the Rust language would have extremely fine-grained isolation boundaries, have no memory leaks, and be safe from a wide range of security threats and memory bugs. Previous efforts towards this end concluded that writing a kernel requires changing Rust. This paper reaches a different conclusion, that no changes to Rust are needed and a kernel can be implemented with a very small amount of unsafe code. It describes how three sample kernel mechanisms–-DMA, USB, and buffer caches–-can be built using these abstractions.

Embedded software is found everywhere from our highly visible mobile devices to the confines of our car in the form of smart sensors. Embedded software companies are under huge pressure to produce safe applications that limit risks, and testing is absolutely critical to alleviate concerns regarding safety and user privacy. This requires using large test suites throughout the development process, increasing time-to-market and ultimately hindering competitivity. Speeding up test execution is, therefore, of paramount importance for embedded software developers. This is traditionally achieved by running, in parallel, multiple tests on large-scale clusters of computers. However, this approach is costly in terms of infrastructure maintenance and energy consumed, and is at times inconvenient as developers have to wait for their tests to be scheduled on a shared resource. We propose to look at exploiting GPUs (Graphics Processing Units) for running embedded software testing. GPUs are readily available in most computers and offer tremendous amounts of parallelism, making them an ideal target for embedded software testing. In this paper, we demonstrate, for the first time, how test executions of embedded C programs can be automatically performed on a GPU, without involving the end user. We take a compiler-assisted approach which automatically compiles the C program into GPU kernels for parallel execution of the input tests. Using this technique, we achieve an average speedup of 16× when compared to CPU execution of input tests across nine programs from an industry standard embedded benchmark suite.

For the last several decades, the rapid development of information technology and computer performance accelerates generation, transportation and accumulation of digital data, it came to be called "Big Data". In this context, researchers and companies are eager to utilize the data to create new values or manage a wide range of issues, and much focus is being placed on "Data Science" to extract useful information (knowledge) from digital data. Data Science has been developed from several independent fields such as Mathematics/Operations Research, Computer Science, Data Engineering, Visualization and Statistics since 1800s. In addition, Artificial Intelligence converges on this stream recent years. On the other hand, the national projects have been established to utilize data for society with concerns surrounding the security and privacy. In this paper, through detailed analysis on history of this field, processes of development and integration among related fields are discussed as well as comparative aspects between Japan and the United States. This paper also includes a brief discussion of future directions.

The risk posed by insider threats has usually been approached by analyzing the behavior of users solely in the cyber domain. In this paper, we show the viability of using physical movement logs, collected via a building access control system, together with an understanding of the layout of the building housing the system's assets, to detect malicious insider behavior that manifests itself in the physical domain. In particular, we propose a systematic framework that uses contextual knowledge about the system and its users, learned from historical data gathered from a building access control system, to select suitable models for representing movement behavior. We then explore the online usage of the learned models, together with knowledge about the layout of the building being monitored, to detect malicious insider behavior. Finally, we show the effectiveness of the developed framework using real-life data traces of user movement in railway transit stations.

Educational games have a potentially significant role to play in the increasing efforts to expand access to computer science education. Computational thinking is an area of particular interest, including the development of problem-solving strategies like divide and conquer. Existing games designed to teach computational thinking generally consist of either open-ended exploration with little direct guidance or a linear series of puzzles with lots of direct guidance, but little exploration. Educational research indicates that the most effective approach may be a hybrid of these two structures. We present Dragon Architect, an educational computational thinking game, and use it as context for a discussion of key open problems in the design of games to teach computational thinking. These problems include how to directly teach computational thinking strategies, how to achieve a balance between exploration and direct guidance, and how to incorporate engaging social features. We also discuss several important design challenges we have encountered during the design of Dragon Architect. We contend the problems we describe are relevant to anyone making educational games or systems that need to teach complex concepts and skills.

The display image on the visual display unit (VDU) can be retrieved from the radiated and conducted emission at some distance with no trace. In this paper, the maximum eavesdropping distance for the unintentional radiation and conduction electromagnetic (EM) signals which contain information has been estimated in theory by considering some realistic parameters. Firstly, the maximum eavesdropping distance for the unintentional EM radiation is estimated based on the reception capacity of a log-periodic antenna which connects to a receiver, the experiment data, the attenuation in free-space and the additional attenuation in the propagation path. And then, based on a multi-conductor transmission model and some experiment results, the maximum eavesdropping distance for the conducted emission is theoretically derived. The estimating results demonstrated that the ITE equipment may also exist threat of the information leakage even if it has met the current EMC requirements.

Here we explore the applicability of traditional sliding window based convolutional neural network (CNN) detection pipeline and region based object detection techniques such as Faster Region-based CNN (R-CNN) and Region-based Fully Convolutional Networks (R-FCN) on the problem of object detection in X-ray security imagery. Within this context, with limited dataset availability, we employ a transfer learning paradigm for network training tackling both single and multiple object detection problems over a number of R-CNN/R-FCN variants. The use of first-stage region proposal within the Faster RCNN and R-FCN provide superior results than traditional sliding window driven CNN (SWCNN) approach. With the use of Faster RCNN with VGG16, pretrained on the ImageNet dataset, we achieve 88.3 mAP for a six object class X-ray detection problem. The use of R-FCN with ResNet-101, yields 96.3 mAP for the two class firearm detection problem requiring 0.1 second computation per image. Overall we illustrate the comparative performance of these techniques as object localization strategies within cluttered X-ray security imagery.

Meta-programs are programs that generate other programs, but in weakly type-safe systems, type-checking a meta-program only establishes its own type safety, and generated programs need additional type-checking after generation. Strong type safety of a meta-program implies type safety of any generated object program, a property with important engineering benefits. Current strongly type-safe systems suffer from expressivity limitations and cannot support many meta-programs found in practice, for example automatic generation of lenses. To overcome this, we move away from the idea of staged meta-programming. Instead, we use an off-the-shelf dependently-typed language as the meta-language and a relatively standard, intrinsically well-typed representation of the object language. We scale this approach to practical meta-programming, by choosing a high-level, explicitly typed intermediate representation as the object language, rather than a surface programming language. We implement our approach as a library for the Glasgow Haskell Compiler (GHC) and evaluate it on several meta-programs, including a deriveLenses meta-program taken from a real-world Haskell lens library. Our evaluation demonstrates expressivity beyond the state of the art and applicability to real settings, at little cost in terms of code size.

Applications for data analysis of biomedical data are complex programs and often consist of multiple components. Re-usage of existing solutions from external code repositories or program libraries is common in algorithm development. To ease reproducibility as well as transfer of algorithms and required components into distributed infrastructures Linux containers are increasingly used in those environments, that are at least partly connected to the internet. However concerns about the untrusted application remain and are of high interest when medical data is processed. Additionally, the portability of the containers needs to be ensured by using only security technologies, that do not require additional kernel modules. In this paper we describe measures and a solution to secure the execution of an example biomedical application for normalization of multidimensional biosignal recordings. This application, the required runtime environment and the security mechanisms are installed in a Docker-based container. A fine-grained restricted environment (sandbox) for the execution of the application and the prevention of unwanted behaviour is created inside the container. The sandbox is based on the filtering of system calls, as they are required to interact with the operating system to access potentially restricted resources e.g. the filesystem or network. Due to the low-level character of system calls, the creation of an adequate rule set for the sandbox is challenging. Therefore the presented solution includes a monitoring component to collect required data for defining the rules for the application sandbox. Performance evaluation of the application execution shows no significant impact of the resulting sandbox, while detailed monitoring may increase runtime up to over 420%.

The challenge of maintaining confidentiality of stored and processed data in a remote database or cloud is quite urgent. Using homomorphic encryption may solve the problem, because it allows to compute some functions over encrypted data without preliminary deciphering of data. Fully homomorphic encryption schemes have a number of limitations such as accumulation of noise and increase of ciphertext extension during performing operations, the range of operations is limited. Nowadays a lot of homomorphic encryption schemes and their modifications have been investigated, so more than 25 reports on homomorphic encryption schemes have already been published on Cryptology ePrint Archive for 2016. We propose an overview of current Fully Homomorphic Encryption Schemes and analyze specific operations for databases which homomorphic cryptosystems allow to perform. We also investigate the possibility of sorting over encrypted data and present our approach to compare data encrypted by Multi-bit FHE scheme.

This paper presents a new fractional-order hidden strange attractor generated by a chaotic system without equilibria. The proposed non-equilibrium fractional-order chaotic system (FOCS) is asymmetric, dissimilar, topologically inequivalent to typical chaotic systems and challenges the conventional notion that the presence of unstable equilibria is mandatory to ensure the existence of chaos. The new fractional-order model displays rich bifurcation undergoing a period doubling route to chaos, where the fractional order α is the bifurcation parameter. Study of the hidden attractor dynamics is carried out with the aid of phase portraits, sensitivity to initial conditions, fractal Lyapunov dimension, maximum Lyapunov exponents spectrum and bifurcation analysis. The minimum commensurate dimension to display chaos is determined. With a view to utilizing it in chaos based cryptology and coding information, a synchronisation control scheme is designed. Finally the theoretical analyses are validated by numerical simulation results which are in good agreement with the former.