Biblio

Intrusion Detection Systems (IDS) have been in existence for many years now, but they fall short in efficiently detecting zero-day attacks. This paper presents an organic combination of Semantic Link Networks (SLN) and dynamic semantic graph generation for the on the fly discovery of zero-day attacks using the Spark Streaming platform for parallel detection. In addition, a minimum redundancy maximum relevance (MRMR) feature selection algorithm is deployed to determine the most discriminating features of the dataset. Compared to previous studies on zero-day attack identification, the described method yields better results due to the semantic learning and reasoning on top of the training data and due to the use of collaborative classification methods. We also verified the scalability of our method in a distributed environment.

Machine learning and data mining algorithms typically assume that the training and testing data are sampled from the same fixed probability distribution; however, this violation is often violated in practice. The field of domain adaptation addresses the situation where this assumption of a fixed probability between the two domains is violated; however, the difference between the two domains (training/source and testing/target) may not be known a priori. There has been a recent thrust in addressing the problem of learning in the presence of an adversary, which we formulate as a problem of domain adaption to build a more robust classifier. This is because the overall security of classifiers and their preprocessing stages have been called into question with the recent findings of adversaries in a learning setting. Adversarial training (and testing) data pose a serious threat to scenarios where an attacker has the opportunity to ``poison'' the training or ``evade'' on the testing data set(s) in order to achieve something that is not in the best interest of the classifier. Recent work has begun to show the impact of adversarial data on several classifiers; however, the impact of the adversary on aspects related to preprocessing of data (i.e., dimensionality reduction or feature selection) has widely been ignored in the revamp of adversarial learning research. Furthermore, variable selection, which is a vital component to any data analysis, has been shown to be particularly susceptible under an attacker that has knowledge of the task. In this work, we explore avenues for learning resilient classification models in the adversarial learning setting by considering the effects of adversarial data and how to mitigate its effects through optimization. Our model forms a single convex optimization problem that uses the labeled training data from the source domain and known- weaknesses of the model for an adversarial component. We benchmark the proposed approach on synthetic data and show the trade-off between classification accuracy and skew-insensitive statistics.

Detecting software security vulnerabilities and distinguishing vulnerable from non-vulnerable code is anything but simple. Most of the time, vulnerabilities remain undisclosed until they are exposed, for instance, by an attack during the software operational phase. Software metrics are widely-used indicators of software quality, but the question is whether they can be used to distinguish vulnerable software units from the non-vulnerable ones during development. In this paper, we perform an exploratory study on software metrics, their interdependency, and their relation with security vulnerabilities. We aim at understanding: i) the correlation between software architectural characteristics, represented in the form of software metrics, and the number of vulnerabilities; and ii) which are the most informative and discriminative metrics that allow identifying vulnerable units of code. To achieve these goals, we use, respectively, correlation coefficients and heuristic search techniques. Our analysis is carried out on a dataset that includes software metrics and reported security vulnerabilities, exposed by security attacks, for all functions, classes, and files of five widely used projects. Results show: i) a strong correlation between several project-level metrics and the number of vulnerabilities, ii) the possibility of using a group of metrics, at both file and function levels, to distinguish vulnerable and non-vulnerable code with a high level of accuracy.

Technological advancement enables the need of internet everywhere. The power industry is not an exception in the technological advancement which makes everything smarter. Smart grid is the advanced version of the traditional grid, which makes the system more efficient and self-healing. Synchrophasor is a device used in smart grids to measure the values of electric waves, voltages and current. The phasor measurement unit produces immense volume of current and voltage data that is used to monitor and control the performance of the grid. These data are huge in size and vulnerable to attacks. Intrusion Detection is a common technique for finding the intrusions in the system. In this paper, a big data framework is designed using various machine learning techniques, and intrusions are detected based on the classifications applied on the synchrophasor dataset. In this approach various machine learning techniques like deep neural networks, support vector machines, random forest, decision trees and naive bayes classifications are done for the synchrophasor dataset and the results are compared using metrics of accuracy, recall, false rate, specificity, and prediction time. Feature selection and dimensionality reduction algorithms are used to reduce the prediction time taken by the proposed approach. This paper uses apache spark as a platform which is suitable for the implementation of Intrusion Detection system in smart grids using big data analytics.

Feature selection is an important step in data analysis to address the curse of dimensionality. Such dimensionality reduction techniques are particularly important when if a classification is required and the model scales in polynomial time with the size of the feature (e.g., some applications include genomics, life sciences, cyber-security, etc.). Feature selection is the process of finding the minimum subset of features that allows for the maximum predictive power. Many of the state-of-the-art information-theoretic feature selection approaches use a greedy forward search; however, there are concerns with the search in regards to the efficiency and optimality. A unified framework was recently presented for information-theoretic feature selection that tied together many of the works in over the past twenty years. The work showed that joint mutual information maximization (JMI) is generally the best options; however, the complexity of greedy search for JMI scales quadratically and it is infeasible on high dimensional datasets. In this contribution, we propose a fast approximation of JMI based on information theory. Our approach takes advantage of decomposing the calculations within JMI to speed up a typical greedy search. We benchmarked the proposed approach against JMI on several UCI datasets, and we demonstrate that the proposed approach returns feature sets that are highly consistent with JMI, while decreasing the run time required to perform feature selection.

Statistical prediction models can be an effective technique to identify vulnerable components in large software projects. Two aspects of vulnerability prediction models have a profound impact on their performance: 1) the features (i.e., the characteristics of the software) that are used as predictors and 2) the way those features are used in the setup of the statistical learning machinery. In a previous work, we compared models based on two different types of features: software metrics and term frequencies (text mining features). In this paper, we broaden the set of models we compare by investigating an array of techniques for the manipulation of said features. These techniques fall under the umbrella of dimensionality reduction and have the potential to improve the ability of a prediction model to localize vulnerabilities. We explore the role of dimensionality reduction through a series of cross-validation and cross-project prediction experiments. Our results show that in the case of software metrics, a dimensionality reduction technique based on confirmatory factor analysis provided an advantage when performing cross-project prediction, yielding the best F-measure for the predictions in five out of six cases. In the case of text mining, feature selection can make the prediction computationally faster, but no dimensionality reduction technique provided any other notable advantage.

Authentication of smartphone users is important because a lot of sensitive data is stored in the smartphone and the smartphone is also used to access various cloud data and services. However, smartphones are easily stolen or co-opted by an attacker. Beyond the initial login, it is highly desirable to re-authenticate end-users who are continuing to access security-critical services and data. Hence, this paper proposes a novel authentication system for implicit, continuous authentication of the smartphone user based on behavioral characteristics, by leveraging the sensors already ubiquitously built into smartphones. We propose novel context-based authentication models to differentiate the legitimate smartphone owner versus other users. We systematically show how to achieve high authentication accuracy with different design alternatives in sensor and feature selection, machine learning techniques, context detection and multiple devices. Our system can achieve excellent authentication performance with 98.1% accuracy with negligible system overhead and less than 2.4% battery consumption.

In this paper, a novel method to do feature selection to detect botnets at their phase of Command and Control (C&C) is presented. A major problem is that researchers have proposed features based on their expertise, but there is no a method to evaluate these features since some of these features could get a lower detection rate than other. To this aim, we find the feature set based on connections of botnets at their phase of C&C, that maximizes the detection rate of these botnets. A Genetic Algorithm (GA) was used to select the set of features that gives the highest detection rate. We used the machine learning algorithm C4.5, this algorithm did the classification between connections belonging or not to a botnet. The datasets used in this paper were extracted from the repositories ISOT and ISCX. Some tests were done to get the best parameters in a GA and the algorithm C4.5. We also performed experiments in order to obtain the best set of features for each botnet analyzed (specific), and for each type of botnet (general) too. The results are shown at the end of the paper, in which a considerable reduction of features and a higher detection rate than the related work presented were obtained.

Heterogeneous face recognition aims to identify or verify person identity by matching facial images of different modalities. In practice, it is known that its performance is highly influenced by modality inconsistency, appearance occlusions, illumination variations and expressions. In this paper, a new method named as ensemble of sparse cross-modal metrics is proposed for tackling these challenging issues. In particular, a weak sparse cross-modal metric learning method is firstly developed to measure distances between samples of two modalities. It learns to adjust rank-one cross-modal metrics to satisfy two sets of triplet based cross-modal distance constraints in a compact form. Meanwhile, a group based feature selection is performed to enforce that features in the same position of two modalities are selected simultaneously. By neglecting features that attribute to "noise" in the face regions (eye glasses, expressions and so on), the performance of learned weak metrics can be markedly improved. Finally, an ensemble framework is incorporated to combine the results of differently learned sparse metrics into a strong one. Extensive experiments on various face datasets demonstrate the benefit of such feature selection especially when heavy occlusions exist. The proposed ensemble metric learning has been shown superiority over several state-of-the-art methods in heterogeneous face recognition.

The ever increasing interest in semantic technologies and the availability of several open knowledge sources have fueled recent progress in the field of recommender systems. In this paper we feed recommender systems with features coming from the Linked Open Data (LOD) cloud - a huge amount of machine-readable knowledge encoded as RDF statements - with the aim of improving recommender systems effectiveness. In order to exploit the natural graph-based structure of RDF data, we study the impact of the knowledge coming from the LOD cloud on the overall performance of a graph-based recommendation algorithm. In more detail, we investigate whether the integration of LOD-based features improves the effectiveness of the algorithm and to what extent the choice of different feature selection techniques influences its performance in terms of accuracy and diversity. The experimental evaluation on two state of the art datasets shows a clear correlation between the feature selection technique and the ability of the algorithm to maximize a specific evaluation metric. Moreover, the graph-based algorithm leveraging LOD-based features is able to overcome several state of the art baselines, such as collaborative filtering and matrix factorization, thus confirming the effectiveness of the proposed approach.

Today's systems produce a rapidly exploding amount of data, and the data further derives more data, forming a complex data propagation network that we call the data's lineage. There are many reasons that users want systems to forget certain data including its lineage. From a privacy perspective, users who become concerned with new privacy risks of a system often want the system to forget their data and lineage. From a security perspective, if an attacker pollutes an anomaly detector by injecting manually crafted data into the training data set, the detector must forget the injected data to regain security. From a usability perspective, a user can remove noise and incorrect entries so that a recommendation engine gives useful recommendations. Therefore, we envision forgetting systems, capable of forgetting certain data and their lineages, completely and quickly. This paper focuses on making learning systems forget, the process of which we call machine unlearning, or simply unlearning. We present a general, efficient unlearning approach by transforming learning algorithms used by a system into a summation form. To forget a training data sample, our approach simply updates a small number of summations – asymptotically faster than retraining from scratch. Our approach is general, because the summation form is from the statistical query learning in which many machine learning algorithms can be implemented. Our approach also applies to all stages of machine learning, including feature selection and modeling. Our evaluation, on four diverse learning systems and real-world workloads, shows that our approach is general, effective, fast, and easy to use.

In this paper, we present the design, architecture, and implementation of a novel analysis engine, called Feature Collection and Correlation Engine (FCCE), that finds correlations across a diverse set of data types spanning over large time windows with very small latency and with minimal access to raw data. FCCE scales well to collecting, extracting, and querying features from geographically distributed large data sets. FCCE has been deployed in a large production network with over 450,000 workstations for 3 years, ingesting more than 2 billion events per day and providing low latency query responses for various analytics. We explore two security analytics use cases to demonstrate how we utilize the deployment of FCCE on large diverse data sets in the cyber security domain: 1) detecting fluxing domain names of potential botnet activity and identifying all the devices in the production network querying these names, and 2) detecting advanced persistent threat infection. Both evaluation results and our experience with real-world applications show that FCCE yields superior performance over existing approaches, and excels in the challenging cyber security domain by correlating multiple features and deriving security intelligence.

An abnormal behavior detection algorithm for surveillance is required to correctly identify the targets as being in a normal or chaotic movement. A model is developed here for this purpose. The uniqueness of this algorithm is the use of foreground detection with Gaussian mixture (FGMM) model before passing the video frames to optical flow model using Lucas-Kanade approach. Information of horizontal and vertical displacements and directions associated with each pixel for object of interest is extracted. These features are then fed to feed forward neural network for classification and simulation. The study is being conducted on the real time videos and some synthesized videos. Accuracy of method has been calculated by using the performance parameters for Neural Networks. In comparison of plain optical flow with this model, improved results have been obtained without noise. Classes are correctly identified with an overall performance equal to 3.4e-02 with & error percentage of 2.5.