Biblio

Advanced Persistent Threats (APTs) are a new breed of internet based smart threats, which can go undetected with the existing state of-the-art internet traffic monitoring and protection systems. With the evolution of internet and cloud computing, a new generation of smart APT attacks has also evolved and signature based threat detection systems are proving to be futile and insufficient. One of the essential strategies in detecting APTs is to continuously monitor and analyze various features of a TCP/IP connection, such as the number of transferred packets, the total count of the bytes exchanged, the duration of the TCP/IP connections, and details of the number of packet flows. The current threat detection approaches make extensive use of machine learning algorithms that utilize statistical and behavioral knowledge of the traffic. However, the performance of these algorithms is far from satisfactory in terms of reducing false negatives and false positives simultaneously. Mostly, current algorithms focus on reducing false positives, only. This paper presents a fractal based anomaly classification mechanism, with the goal of reducing both false positives and false negatives, simultaneously. A comparison of the proposed fractal based method with a traditional Euclidean based machine learning algorithm (k-NN) shows that the proposed method significantly outperforms the traditional approach by reducing false positive and false negative rates, simultaneously, while improving the overall classification rates.

Complex networks usually expose community structure with groups of nodes sharing many links with the other nodes in the same group and relatively few with the nodes of the rest. This feature captures valuable information about the organization and even the evolution of the network. Over the last decade, a great number of algorithms for community detection have been proposed to deal with the increasingly complex networks. However, the problem of doing this in a private manner is rarely considered. In this paper, we solve this problem under differential privacy, a prominent privacy concept for releasing private data. We analyze the major challenges behind the problem and propose several schemes to tackle them from two perspectives: input perturbation and algorithm perturbation. We choose Louvain method as the back-end community detection for input perturbation schemes and propose the method LouvainDP which runs Louvain algorithm on a noisy super-graph. For algorithm perturbation, we design ModDivisive using exponential mechanism with the modularity as the score. We have thoroughly evaluated our techniques on real graphs of different sizes and verified that ModDivisive steadily gives the best modularity and avg.F1Score on large graphs while LouvainDP outperforms the remaining input perturbation competitors in certain settings.

The safety of information inside of cloud networks is of interest to the network administrators. In a new insider attack, inside attackers merge confidential information with videos using digital video steganography. The video can be uploaded to video websites, where the information can be distributed online, where it can cost firms millions in damages. Standard behavior based exfiltration detection does not always prevent these attacks. This form of steganography is almost invisible. Existing compressed video steganalysis only detects small-payload watermarks. We develop such a strategy using distributed algorithms and classify videos, then compare existing algorithms to new ones. We find our approach improves on behavior based exfiltration detection, and on the existing online video steganalysis.

Traditional sensitive data disclosure analysis faces two challenges: to identify sensitive data that is not generated by specific API calls, and to report the potential disclosures when the disclosed data is recognized as sensitive only after the sink operations. We address these issues by developing BidText, a novel static technique to detect sensitive data disclosures. BidText formulates the problem as a type system, in which variables are typed with the text labels that they encounter (e.g., during key-value pair operations). The type system features a novel bi-directional propagation technique that propagates the variable label sets through forward and backward data-flow. A data disclosure is reported if a parameter at a sink point is typed with a sensitive text label. BidText is evaluated on 10,000 Android apps. It reports 4,406 apps that have sensitive data disclosures, with 4,263 apps having log based disclosures and 1,688 having disclosures due to other sinks such as HTTP requests. Existing techniques can only report 64.0% of what BidText reports. And manual inspection shows that the false positive rate for BidText is 10%.

A normal computer infected with malware is difficult to detect. There have been several approaches in the last years which analyze the behavior of malware and obtain good results. The malware traffic may be detected, but it is very common to miss-detect normal traffic as malicious and generate false positives. This is specially the case when the methods are tested in real and large networks. The detection errors are generated due to the malware changing and rapidly adapting its domains and patterns to mimic normal connections. To better detect malware infections and separate them from normal traffic we propose to detect the behavior of the group of connections generated by the malware. It is known that malware usually generates various related connections simultaneously and therefore it shows a group pattern. Based on previous experiments, this paper suggests that the behavior of a group of connections can be modelled as a directed cyclic graph with special properties, such as its internal patterns, relationships, frequencies and sequences of connections. By training the group models on known traffic it may be possible to better distinguish between a malware connection and a normal connection.

The Domain Name System (DNS) is a critically fundamental element in the internet technology as it translates domain names into corresponding IP addresses. The DNS queries and responses are UDP (User Datagram Protocol) based. DNS name servers are constantly facing threats of DNS amplification attacks. DNS amplification attack is one of the major Distributed Denial of Service (DDoS) attacks, in DNS. The DNS amplification attack victimized huge business and financial companies and organizations by giving disturbance to the customers. In this paper, a mechanism is proposed to detect such attacks coming from the compromised machines. We analysed DNS traffic packet comparatively based on the Machine Learning Classification algorithms such as Decision Tree (TREE), Multi Layer Perceptron (MLP), Naïve Bayes (NB) and Support Vector Machine (SVM) to classify the DNS traffics into normal and abnormal. In this approach attribute selection algorithms such as Information Gain, Gain Ratio and Chi Square are used to achieve optimal feature subset. In the experimental result it shows that the Decision Tree achieved 99.3% accuracy. This model gives highest accuracy and performance as compared to other Machine Learning algorithms.

Web applications are used on a large scale worldwide, which handles sensitive personal data of users. With web application that maintains data ranging from as simple as telephone number to as important as bank account information, security is a prime point of concern. With hackers aimed to breakthrough this security using various attacks, we are focusing on SQL injection attacks and XSS attacks. SQL injection attack is very common attack that manipulates the data passing through web application to the database servers through web servers in such a way that it alters or reveals database contents. While Cross Site Scripting (XSS) attacks focuses more on view of the web application and tries to trick users that leads to security breach. We are considering three tier web applications with static and dynamic behavior, for security. Static and dynamic mapping model is created to detect anomalies in the class of SQL Injection and XSS attacks.

This paper describes an approach for detecting the presence of domain name system (DNS) tunnels in network traffic. DNS tunneling is a common technique hackers use to establish command and control nodes and to exfiltrate data from networks. To generate the training data sufficient to build models to detect DNS tunneling activity, a penetration testing effort was employed. We extracted features from this data and trained random forest classifiers to distinguish normal DNS activity from tunneling activity. The classifiers successfully detected the presence of tunnels we trained on, and four other types of tunnels that were not a part of the training set.

Steganography is the science of hiding data within data. Either for the good purpose of secret communication or for the bad intention of leaking sensitive confidential data or embedding malicious code or URL. However, many different carrier file formats can be used to hide these data (network, audio, image..etc) but the most common steganography carrier is embedding secret data within images as it is considered to be the best and easiest way to hide all types of files (secret files) within an image using different formats (another image, text, video, virus, URL..etc). To the human eye, the changes in the image appearance with the hidden data can be imperceptible. In fact, images can be more than what we see with our eyes. Therefore, many solutions where proposed to help in detecting these hidden data but each solution have their own strong and weak points either by the limitation of resolving one type of image along with specific hiding technique and or most likely without extracting the hidden data. This paper intends to propose a novel detection approach that will concentrate on detecting any kind of hidden URL in all types of images and extract the hidden URL from the carrier image that used the LSB least significant bit hiding technique.

An increasingly important concern of software engineers is handling uncertainty at runtime. Over the last decade researchers have applied architecture-based self-adaptation approaches to address this concern. However, providing guarantees required by current software systems has shown to be challenging with these approaches. To tackle this challenge, we study the application of control theory to realize self-adaptation and develop novel control-based adaptation mechanisms that guarantee desired system properties. Results are validated on systems with strict requirements.

Research towards my dissertation has involved a series of perceptual and accessibility-focused studies concerned with the use of tactile cues for spatial and situational awareness, displayed through head-mounted wearables. These studies were informed by an initial participatory design study of mobile technology multitasking and tactile interaction habits. This research has yielded a number of actionable conclusions regarding the development of tactile interfaces for the head, and endeavors to provide greater insight into the design of advanced tactile alerting for contextual and spatial understanding in assistive applications (e.g. for individuals who are blind or those encountering situational impairments), as well as guidance for developers regarding assessment of interaction between under-utilized sensory modalities and underlying perceptual and cognitive processes.

Smart Transportation applications by nature are examples of Vehicular Ad-hoc Network (VANETs) applications where mobile vehicles, roadside units and transportation infrastructure interplay with one another to provide value added services. While there are abundant researches that focused on the communication aspect of such Mobile Ad-hoc Networks, there are few research bodies that target the development of VANET applications. Among the popular VANET applications, a dominant direction is to leverage Cloud infrastructure to execute and deliver applications and services. Recent studies showed that Cloud Computing is not sufficient for many VANET applications due to the mobility of vehicles and the latency sensitive requirements they impose. To this end, Fog Computing has been proposed to leverage computation infrastructure that is closer to the network edge to compliment Cloud Computing in providing latency-sensitive applications and services. However, applications development in Fog environment is much more challenging than in the Cloud due to the distributed nature of Fog systems. In this paper, we investigate how Smart Transportation applications are developed following Fog Computing approach, their challenges and possible mitigation from the state of the arts.

Collaborative filtering plays an essential role in a recommender system, which recommends a list of items to a user by learning behavior patterns from user rating matrix. However, if an attacker has some auxiliary knowledge about a user purchase history, he/she can infer more information about this user. This brings great threats to user privacy. Some methods adopt differential privacy algorithms in collaborative filtering by adding noises to a rating matrix. Although they provide theoretically private results, the influence on recommendation accuracy are not discussed. In this paper, we solve the privacy problem in recommender system in a different way by applying the differential privacy method into the procedure of recommendation. We design two differentially private recommender algorithms with sampling, named Differentially Private Item Based Recommendation with sampling (DP-IR for short) and Differentially Private User Based Recommendation with sampling(DP-UR for short). Both algorithms are based on the exponential mechanism with a carefully designed quality function. Theoretical analyses on privacy of these algorithms are presented. We also investigate the accuracy of the proposed method and give theoretical results. Experiments are performed on real datasets to verify our methods.

Location entropy (LE) is a popular metric for measuring the popularity of various locations (e.g., points-of-interest). Unlike other metrics computed from only the number of (unique) visits to a location, namely frequency, LE also captures the diversity of the users' visits, and is thus more accurate than other metrics. Current solutions for computing LE require full access to the past visits of users to locations, which poses privacy threats. This paper discusses, for the first time, the problem of perturbing location entropy for a set of locations according to differential privacy. The problem is challenging because removing a single user from the dataset will impact multiple records of the database; i.e., all the visits made by that user to various locations. Towards this end, we first derive non-trivial, tight bounds for both local and global sensitivity of LE, and show that to satisfy ε-differential privacy, a large amount of noise must be introduced, rendering the published results useless. Hence, we propose a thresholding technique to limit the number of users' visits, which significantly reduces the perturbation error but introduces an approximation error. To achieve better utility, we extend the technique by adopting two weaker notions of privacy: smooth sensitivity (slightly weaker) and crowd-blending (strictly weaker). Extensive experiments on synthetic and real-world datasets show that our proposed techniques preserve original data distribution without compromising location privacy.

Social Networking is fundamentally shifting the way we communicate, sharing idea and form opinions. All people try to use social media for there need, people from every age group are involved in social media site or e-commerce site. Nowadays almost every illegal activity is happened using the social network and instant messages. It means that present system is not capable to found all suspicious words. In this paper, we provided a brief description of problem and review on the different framework developed so far. Propose a better system which can be indentify criminal activity through social networking more efficiently. Use Ontology Based Information Extraction (OBIE) technique to identify domain of word and Association Rule mining to generate rules. Heuristic method checks in user database for malicious users according to predefine elements and Naïve Bayes method is use to identify the context behind the message or post. The experimental result is used for further action on victim by cyber crime department.

Hashing based methods have attracted considerable attention for efficient cross-modal retrieval on large-scale multimedia data. The core problem of cross-modal hashing is how to effectively integrate heterogeneous features from different modalities to learn hash functions using available supervising information, e.g., class labels. Existing hashing based methods generally project heterogeneous features to a common space for hash codes generation, and the supervising information is incrementally used for improving performance. However, these methods may produce ineffective hash codes, due to the failure to explore the discriminative property of supervising information and to effectively bridge the semantic gap between different modalities. To address these challenges, we propose a novel hashing based method in a linear classification framework, in which the proposed method learns modality-specific hash functions for generating unified binary codes, and these binary codes are viewed as representative features for discriminative classification with class labels. An effective optimization algorithm is developed for the proposed method to jointly learn the modality-specific hash function, the unified binary codes and a linear classifier. Extensive experiments on three benchmark datasets highlight the advantage of the proposed method and show that it achieves the state-of-the-art performance.

Modelling network traffic is gaining importance to counter modern security threats of ever increasing sophistication. It is though surprisingly difficult and costly to construct reliable classifiers on top of telemetry data due to the variety and complexity of signals that no human can manage to interpret in full. Obtaining training data with sufficiently large and variable body of labels can thus be seen as a prohibitive problem. The goal of this work is to detect infected computers by observing their HTTP(S) traffic collected from network sensors, which are typically proxy servers or network firewalls, while relying on only minimal human input in the model training phase. We propose a discriminative model that makes decisions based on a computer's all traffic observed during a predefined time window (5 minutes in our case). The model is trained on traffic samples collected over equally-sized time windows for a large number of computers, where the only labels needed are (human) verdicts about the computer as a whole (presumed infected vs. presumed clean). As part of training, the model itself learns discriminative patterns in traffic targeted to individual servers and constructs the final high-level classifier on top of them. We show the classifier to perform with very high precision, and demonstrate that the learned traffic patterns can be interpreted as Indicators of Compromise. We implement the discriminative model as a neural network with special structure reflecting two stacked multi instance problems. The main advantages of the proposed configuration include not only improved accuracy and ability to learn from gross labels, but also automatic learning of server types (together with their detectors) that are typically visited by infected computers.

IEEE 802.15.4 is widely used as lower layers for not only wellknown wireless communication standards such as ZigBee, 6LoWPAN, and WirelessHART, but also customized protocols developed by manufacturers, particularly for various Internet of Things (IoT) devices. Customized protocols are not usually publicly disclosed nor standardized. Moreover, unlike textual protocols (e.g., HTTP, SMTP, POP3.), customized protocols for IoT devices provide no clues such as strings or keywords that are useful for analysis. Instead, they use bits or bytes to represent header and body information in order to save power and bandwidth. On the other hand, they often do not employ encryption, fragmentation, or authentication to save cost and effort in implementations. In other words, their security relies only on the confidentiality of the protocol itself. In this paper, we introduce a novel methodology to analyze and reconstruct unknown wireless customized protocols over IEEE 802.15.4. Based on this methodology, we develop an automatic analysis and spoofing tool called WPAN automatic spoofer (WASp) that can be used to understand and reconstruct customized protocols to byte-level accuracy, and to generate packets that can be used for verification of analysis results or spoofing attacks. The methodology consists of four phases: packet collection, packet grouping, protocol analysis, and packet generation. Except for the packet collection step, all steps are fully automated. Although the use of customized protocols is also unknown before the collecting phase, we choose two real-world target systems for evaluation: the smart plug system and platform screen door (PSD) to evaluate our methodology and WASp. In the evaluation, 7,299 and 217 packets are used as datasets for both target systems, respectively. As a result, on average, WASp is found to reduce entropy of legitimate message space by 93.77% and 88.11% for customized protocols used in smart plug and PSD systems, respectively. In addition, on average, 48.19% of automatically generated packets are successfully spoofed for the first target systems.

We propose a distributed and adaptive trust evaluation algorithm (DATEA) to calculate the trust between nodes. First, calculate the communication trust by using the number of data packets between nodes, and predict the trust based on the trend of this value, calculate the comprehensive trust by combining the history trust with the predict value; calculate the energy trust based on the residual energy of nodes; calculate the direct trust by using the communication trust and energy trust. Second, calculate the recommendation trust based on the recommendation reliability and the recommendation familiarity; put forward the adaptively weighting method, and calculate the integrate direct trust by combining the direct trust with recommendation trust. Third, according to the integrate direct trust, considering the factor of trust propagation distance, the indirect trust between nodes is calculated. Simulation experiments show that the proposed algorithm can effectively avoid the attacks of malicious nodes, besides, the calculated direct trust and indirect trust about normal nodes are more conformable to the actual situation.

Balanced partitioning is often a crucial first step in solving large-scale graph optimization problems: in some cases, a big graph is chopped into pieces that fit on one machine to be processed independently before stitching the results together, leading to certain suboptimality from the interaction among different pieces. In other cases, links between different parts may show up in the running time and/or network communications cost, hence the desire to have small cut size. We study a distributed balanced partitioning problem where the goal is to partition the vertices of a given graph into k pieces, minimizing the total cut size. Our algorithm is composed of a few steps that are easily implementable in distributed computation frameworks, e.g., MapReduce. The algorithm first embeds nodes of the graph onto a line, and then processes nodes in a distributed manner guided by the linear embedding order. We examine various ways to find the first embedding, e.g., via a hierarchical clustering or Hilbert curves. Then we apply four different techniques such as local swaps, minimum cuts on partition boundaries, as well as contraction and dynamic programming. Our empirical study compares the above techniques with each other, and to previous work in distributed algorithms, e.g., a label propagation method, FENNEL and Spinner. We report our results both on a private map graph and several public social networks, and show that our results beat previous distributed algorithms: we notice, e.g., 15-25% reduction in cut size over [UB13]. We also observe that our algorithms allow for scalable distributed implementation for any number of partitions. Finally, we apply our techniques for the Google Maps Driving Directions to minimize the number of multi-shard queries with the goal of saving in CPU usage. During live experiments, we observe an ≈ 40% drop in the number of multi-shard queries when comparing our method with a standard geography-based method.

In this paper, the design of an event-driven middleware for general purpose services in smart grid (SG) is presented. The main purpose is to provide a peer-to-peer distributed software infrastructure to allow the access of new multiple and authorized actors to SGs information in order to provide new services. To achieve this, the proposed middleware has been designed to be: 1) event-based; 2) reliable; 3) secure from malicious information and communication technology attacks; and 4) to enable hardware independent interoperability between heterogeneous technologies. To demonstrate practical deployment, a numerical case study applied to the whole U.K. distribution network is presented, and the capabilities of the proposed infrastructure are discussed.

Preprocessors support the diversification of software products with \#ifdefs, but also require additional effort from developers to maintain and understand variable code. We conjecture that \#ifdefs cause developers to produce more vulnerable code because they are required to reason about multiple features simultaneously and maintain complex mental models of dependencies of configurable code. We extracted a variational call graph across all configurations of the Linux kernel, and used configuration complexity metrics to compare vulnerable and non-vulnerable functions considering their vulnerability history. Our goal was to learn about whether we can observe a measurable influence of configuration complexity on the occurrence of vulnerabilities. Our results suggest, among others, that vulnerable functions have higher variability than non-vulnerable ones and are also constrained by fewer configuration options. This suggests that developers are inclined to notice functions appear in frequently-compiled product variants. We aim to raise developers' awareness to address variability more systematically, since configuration complexity is an important, but often ignored aspect of software product lines.

Preprocessors support the diversification of software products with \#ifdefs, but also require additional effort from developers to maintain and understand variable code. We conjecture that \#ifdefs cause developers to produce more vulnerable code because they are required to reason about multiple features simultaneously and maintain complex mental models of dependencies of configurable code. We extracted a variational call graph across all configurations of the Linux kernel, and used configuration complexity metrics to compare vulnerable and non-vulnerable functions considering their vulnerability history. Our goal was to learn about whether we can observe a measurable influence of configuration complexity on the occurrence of vulnerabilities. Our results suggest, among others, that vulnerable functions have higher variability than non-vulnerable ones and are also constrained by fewer configuration options. This suggests that developers are inclined to notice functions appear in frequently-compiled product variants. We aim to raise developers' awareness to address variability more systematically, since configuration complexity is an important, but often ignored aspect of software product lines.

There are many techniques to improve software quality. One is using automatic static analysis tools. We have observed, however, that despite the low-cost help they offer, these tools are underused and often discourage beginners. There is evidence that personality traits influence the perceived usability of a software. Thus, to support beginners better, we need to understand how the workflow of people with different prevalent personality traits using these tools varies. For this purpose, we observed users' solution strategies and correlated them with their prevalent personality traits in an exploratory study with student participants within a controlled experiment. We gathered data by screen capturing and chat protocols as well as a Big Five personality traits test. We found strong correlations between particular personality traits and different strategies of removing the findings of static code analysis as well as between personality and tool utilization. Based on that, we offer take-away improvement suggestions. Our results imply that developers should be aware of these solution strategies and use this information to build tools that are more appealing to people with different prevalent personality traits.

We present a study examining the impact of physical and cognitive challenge on reported immersion for a mixed reality game called Beach Pong. Contrary to prior findings for desktop games, we find significantly higher reported immersion among players who engage physically, regardless of their actual game performance. Building a mental map of the real, virtual, and sensed world is a cognitive challenge for novices, and this appears to influence immersion: in our study, participants who actively attended to both physical and virtual game elements reported higher immersion levels than those who attended mainly or exclusively to virtual elements. Without an integrated mental map, in-game cognitive challenges were ignored or offloaded to motor response when possible in order to achieve the minimum required goals of the game. From our results we propose a model of immersion in mixed reality gaming that is useful for designers and researchers in this space.