Biblio

Research has observed context factors like occasion and time as influential factors for predicting whether or not to share a location with online friends. In other domains like social networks, personality was also found to play an important role. Furthermore, users are seeking a fine-grained disclosement policy that also allows them to display an obfuscated location, like the center of the current city, to some of their friends. In this paper, we observe which context factors and personality measures can be used to predict the correct privacy level out of seven privacy levels, which include obfuscation levels like center of the street or current city. Our results show that a prediction is possible with a precision 20% better than a constant value. We will give design indications to determine which context factors should be recorded, and how much the precision can be increased if personality and privacy measures are recorded using either a questionnaire or automated text analysis.

Phishing attacks are one of the most common and least defended security threats today. We present an approach which uses natural language processing techniques to analyze text and detect inappropriate statements which are indicative of phishing attacks. Our approach is novel compared to previous work because it focuses on the natural language text contained in the attack, performing semantic analysis of the text to detect malicious intent. To demonstrate the effectiveness of our approach, we have evaluated it using a large benchmark set of phishing emails.

It is time-consuming and labor-intensive to learn and locate the correct API for programming tasks. Thus, it is beneficial to perform API recommendation automatically. The graph-based statistical model has been shown to recommend top-10 API candidates effectively. It falls short, however, in accurately recommending an actual top-1 API. To address this weakness, we propose RecRank, an approach and tool that applies a novel ranking-based discriminative approach leveraging API usage path features to improve top-1 API recommendation. Empirical evaluation on a large corpus of (1385+8) open source projects shows that RecRank significantly improves top-1 API recommendation accuracy and mean reciprocal rank when compared to state-of-the-art API recommendation approaches.

The advent of machine learning has made it a popular tool in various areas. It has also been applied in network intrusion detection. However, machine learning hasn't been sufficiently explored in the cyberphysical domains such as smart grids. This is because a lot of factors weigh in while using these tools. This paper is about intrusion detection in smart grids and how some machine learning techniques can help achieve this goal. It considers the problems of feature and classifier selection along with other data ambiguities. The goal is to apply the machine learning ensemble classifiers on the smart grid traffic and evaluate if these methods can detect anomalies in the system.

Despite the popularity of wireless sensor networks (WSNs) in a wide range of applications, security problems associated with them have not been completely resolved. Middleware is generally introduced as an intermediate layer between WSNs and the end user to resolve some limitations, but most of the existing middleware is unable to protect data from malicious and unknown attacks during transmission. This paper introduces an intelligent middleware based on an unsupervised learning technique called Generative Adversarial Networks (GANs) algorithm. GANs contain two networks: a generator (G) network and a detector (D) network. The G creates fake data similar to the real samples and combines it with real data from the sensors to confuse the attacker. The D contains multi-layers that have the ability to differentiate between real and fake data. The output intended for this algorithm shows an actual interpretation of the data that is securely communicated through the WSN. The framework is implemented in Python with experiments performed using Keras. Results illustrate that the suggested algorithm not only improves the accuracy of the data but also enhances its security by protecting data from adversaries. Data transmission from the WSN to the end user then becomes much more secure and accurate compared to conventional techniques.

Software defined networking is a concept proposed to replace traditional networks by separating control plane and data plane. It makes the network more programmable and manageable. As there is a single point of control of the network, it is more vulnerable to intrusion. The idea is to train the network controller by machine learning algorithms to let it make the intelligent decisions automatically. In this paper, we have discussed our approach to make software defined networking more secure from various malicious attacks by making it capable of detecting and preventing such attacks.

Nowadays, Volunteered Geographic Information (VGI) is commonly used in research and practical applications. However, the quality assurance of such a geographic data remains a problem. In this study we use machine learning and natural language processing to improve record retrieval by category (e.g. restaurant, museum, etc.) from Wikimapia Points of Interest data. We use textual information contained in VGI records to evaluate its ability to determine the category label. The performance of the trained classifier is evaluated on the complete dataset and then is compared with its performance on regional subsets. Preliminary analysis shows significant difference in the classifier performance across the regions. Such geographic differences will have a significant effect on data enrichment efforts such as labeling entities with missing categories.

The growth of malware poses a major threat to internet users, governments, and businesses around the world. One of the major types of malware, ransomware, encrypts a user's sensitive information and only returns the original files to the user after a ransom is paid. As malware developers shift the delivery of their product from HTTP to HTTPS to protect themselves from payload inspection, we can no longer rely on deep packet inspection to extract features for malware identification. Toward this goal, we propose a solution leveraging a recent trend in networking hardware, that is programmable forwarding engines (PFEs). PFEs allow collection of per-packet, network monitoring data at high rates. We use this data to monitor the network traffic between an infected computer and the command and control (C&C) server. We extract high-level flow features from this traffic and use this data for ransomware classification. We write a stream processor and use a random forest, binary classifier to utilizes these rich flow records in fingerprinting malicious, network activity without the requirement of deep packet inspection. Our classification model achieves a detection rate in excess of 0.86, while maintaining a false negative rate under 0.11. Our results suggest that a flow-based fingerprinting method is feasible and accurate enough to catch ransomware before encryption.

There is growing recognition that machine learning (ML) exposes new security and privacy vulnerabilities in software systems, yet the technical community's understanding of the nature and extent of these vulnerabilities remains limited but expanding. In this talk, we explore the threat model space of ML algorithms through the lens of Saltzer and Schroeder's principles for the design of secure computer systems. This characterization of the threat space prompts an investigation of current and future research directions. We structure our discussion around three of these directions, which we believe are likely to lead to significant progress. The first seeks to design mechanisms for assembling reliable records of compromise that would help understand the degree to which vulnerabilities are exploited by adversaries, as well as favor psychological acceptability of machine learning applications. The second encompasses a spectrum of approaches to input verification and mediation, which is a prerequisite to enable fail-safe defaults in machine learning systems. The third pursues formal frameworks for security and privacy in machine learning, which we argue should strive to align machine learning goals such as generalization with security and privacy desirata like robustness or privacy. Key insights resulting from these three directions pursued both in the ML and security communities are identified and the effectiveness of approaches are related to structural elements of ML algorithms and the data used to train them. We conclude by systematizing best practices in our growing community.

Security Requirements classification is an important area to the Software Engineering community in order to build software that is secure, robust and able to withstand attacks. This classification facilitates proper analysis of security requirements so that adequate security mechanisms are incorporated in the development process. Machine Learning techniques have been used in Security Requirements classification to aid in the process that lead to ensuring that correct security mechanisms are designed corresponding to the Security Requirements classifications made to eliminate the risk of security being incorporated in the late stages of development. However, these Machine Learning techniques have been found to have problems including, handcrafting of features, overfitting and failure to perform well with high dimensional data. In this paper we explore Natural Language Processing and Deep Learning to determine if this can be applied to Security Requirements classification.

Conversational agents have very much become part of our lives since the renaissance of neural network based "neural conversational agents". Previously used manually annotated and rule based methods lacked the scalability and generalization capabilities of the neural conversational agents. A neural conversational agent has two parts: at one end an encoder understands the question while the other end a decoder prepares and outputs the corresponding answer to the question asked. Both the parts are typically designed using recurrent neural network and its variants and trained in an end-to-end fashion. Although conversation agents for other languages have been developed, Urdu language has seen very less progress in building of conversational agents. Especially recent state of the art neural network based techniques have not been explored yet. In this paper, we design an attention driven deep encoder-decoder based neural conversational agent for Urdu language. Overall, we make following contributions we (i) create a dataset of 5000 question-answer pairs, and (ii) present a new deep encoder-decoder based conversational agent for Urdu language. For our work, we limit the knowledge base of our agent to general knowledge regarding Pakistan. Our best model has the BLEU score of 58 and gives syntactically and semantically correct answers in majority of the cases.

Software Defined Network (SDN) is a revolutionary idea to realize software-driven network with the separation of control and data planes. In essence, SDN addresses the problems faced by the traditional network architecture; however, it may as well expose the network to new attacks. Among other attacks, distributed denial of service (DDoS) attacks are hard to contain in such software-based networks. Existing DDoS mitigation techniques either lack in performance or jeopardize the accuracy of the attack detection. To fill the voids, we propose in this paper a machine learning-based DDoS mitigation technique for SDN. First, we create a model for DDoS detection in SDN using NSL-KDD dataset and then after training the model on this dataset, we use real DDoS attacks to assess our proposed model. Obtained results show that the proposed technique equates favorably to the current techniques with increased performance and accuracy.

We propose a novel framework for outsourced private function evaluation with privacy policy enforcement (OPFE-PPE). Suppose an evaluator evaluates a function with private data contributed by a data contributor, and a client obtains the result of the evaluation. OPFE-PPE enables a data contributor to enforce two different kinds of privacy policies to the process of function evaluation: evaluator policy and client policy. An evaluator policy restricts entities that can conduct function evaluation with the data. A client policy restricts entities that can obtain the result of function evaluation. We demonstrate our construction with three applications: personalized medication, genetic epidemiology, and prediction by machine learning. Experimental results show that the overhead caused by enforcing the two privacy policies is less than 10% compared to function evaluation by homomorphic encryption without any privacy policy enforcement.

It is clear that cyber-attacks are a danger that must be addressed with great resolve, as they threaten the information infrastructure upon which we all depend. Many studies have been published expressing varying levels of success with machine learning approaches to combating cyber-attacks, but many modern studies still focus on training and evaluating with very outdated datasets containing old attacks that are no longer a threat, and also lack data on new attacks. Recent datasets like UNSW-NB15 and SANTA have been produced to address this problem. Even so, these modern datasets suffer from class imbalance, which reduces the efficacy of predictive models trained using these datasets. Herein we evaluate several pre-processing methods for addressing the class imbalance problem; using several of the most popular machine learning algorithms and a variant of UNSW-NB15 based upon the attributes from the SANTA dataset.

Website defacements can inflict significant harm on the website owner through the loss of reputation, the loss of money, or the leakage of information. Due to the complexity and diversity of all kinds of web application systems, especially a lack of necessary security maintenance, website defacements increased year by year. In this paper, we focus on detecting whether the website has been defaced by extracting website features and website embedded trojan features. We use three kinds of classification learning algorithms which include Gradient Boosting Decision Tree (GBDT), Random Forest (RF) and Support Vector Machine (SVM) to do the classification experiments, and experimental results show that Support Vector Machine classifier performed better than two other classifiers. It can achieve an overall accuracy of 95%-96% in detecting website defacements.

As a new research hotspot in the field of artificial intelligence, deep reinforcement learning (DRL) has achieved certain success in various fields such as robot control, computer vision, natural language processing and so on. At the same time, the possibility of its application being attacked and whether it have a strong resistance to strike has also become a hot topic in recent years. Therefore, we select the representative Deep Q Network (DQN) algorithm in deep reinforcement learning, and use the robotic automatic pathfinding application as a countermeasure application scenario for the first time, and attack DQN algorithm against the vulnerability of the adversarial samples. In this paper, we first use DQN to find the optimal path, and analyze the rules of DQN pathfinding. Then, we propose a method that can effectively find vulnerable points towards White-Box Q table variation in DQN pathfinding training. Finally, we build a simulation environment as a basic experimental platform to test our method, through multiple experiments, we can successfully find the adversarial examples and the experimental results show that the supervised method we proposed is effective.

With the growing complexity of environments in which systems are expected to operate, adaptive human-machine teaming (HMT) has emerged as a key area of research. While human teams have been extensively studied in the psychological and training literature, and agent teams have been investigated in the artificial intelligence research community, the commitment to research in HMT is relatively new and fueled by several technological advances such as electrophysiological sensors, cognitive modeling, machine learning, and adaptive/adaptable human-machine systems. This paper presents an architectural framework for investigating HMT options in various simulated operational contexts including responding to systemic failures and external disruptions. The paper specifically discusses new and novel roles for machines made possible by new technology and offers key insights into adaptive human-machine teams. Landed aircraft perimeter security is used as an illustrative example of an adaptive cyber-physical-human system (CPHS). This example is used to illuminate the use of the HMT framework in identifying the different human and machine roles involved in this scenario. The framework is domain-independent and can be applied to both defense and civilian adaptive HMT. The paper concludes with recommendations for advancing the state-of-the-art in HMT. 

The term "artificial intelligence" is a buzzword today and is heavily used to market products, services, research, conferences, and more. It is scientifically disputed which types of products and services do actually qualify as "artificial intelligence" versus simply advanced computer technologies mimicking aspects of natural intelligence. Yet it is undisputed that, despite often inflationary use of the term, there are mainstream products and services today that for decades were only thought to be science fiction. They range from industrial automation, to self-driving cars, robotics, and consumer electronics for smart homes, workspaces, education, and many more contexts. Several technological advances enable what is commonly referred to as "artificial intelligence". It includes connected computers and the Internet of Things (IoT), open and big data, low cost computing and storage, and many more. Yet regardless of the definition of the term artificial intelligence, technological advancements in this area provide immense potential, especially for people with disabilities. In this paper we explore some of these potential in the context of web accessibility. We review some existing products and services, and their support for web accessibility. We propose accessibility conformance evaluation as one potential way forward, to accelerate the uptake of artificial intelligence, to improve web accessibility.

Telephone spam has become an increasingly prevalent problem in many countries all over the world. For example, the US Federal Trade Commission's (FTC) National Do Not Call Registry's number of cumulative complaints of spam/scam calls reached 30.9 million submissions in 2016. Naturally, telephone carriers can play an important role in the fight against spam. However, due to the extremely large volume of calls that transit across large carrier networks, it is challenging to mine their vast amounts of call detail records (CDRs) to accurately detect and block spam phone calls. This is because CDRs only contain high-level metadata (e.g., source and destination numbers, call start time, call duration, etc.) related to each phone calls. In addition, ground truth about both benign and spam-related phone numbers is often very scarce (only a tiny fraction of all phone numbers can be labeled). More importantly, telephone carriers are extremely sensitive to false positives, as they need to avoid blocking any non-spam calls, making the detection of spam-related numbers even more challenging. In this paper, we present a novel detection system that aims to discover telephone numbers involved in spam campaigns. Given a small seed of known spam phone numbers, our system uses a combination of unsupervised and supervised machine learning methods to mine new, previously unknown spam numbers from large datasets of call detail records (CDRs). Our objective is not to detect all possible spam phone calls crossing a carrier's network, but rather to expand the list of known spam numbers while aiming for zero false positives, so that the newly discovered numbers may be added to a phone blacklist, for example. To evaluate our system, we have conducted experiments over a large dataset of real-world CDRs provided by a leading telephony provider in China, while tuning the system to produce no false positives. The experimental results show that our system is able to greatly expand on the initial seed of known spam numbers by up to about 250%.

Although various techniques have been proposed to generate adversarial samples for white-box attacks on text, little attention has been paid to a black-box attack, which is a more realistic scenario. In this paper, we present a novel algorithm, DeepWordBug, to effectively generate small text perturbations in a black-box setting that forces a deep-learning classifier to misclassify a text input. We develop novel scoring strategies to find the most important words to modify such that the deep classifier makes a wrong prediction. Simple character-level transformations are applied to the highest-ranked words in order to minimize the edit distance of the perturbation. We evaluated DeepWordBug on two real-world text datasets: Enron spam emails and IMDB movie reviews. Our experimental results indicate that DeepWordBug can reduce the classification accuracy from 99% to 40% on Enron and from 87% to 26% on IMDB. Our results strongly demonstrate that the generated adversarial sequences from a deep-learning model can similarly evade other deep models.

The explosive proliferation of Internet of Things (IoT) devices is generating an incomprehensible amount of data. Machine learning plays an imperative role in aggregating this data and extracting valuable information for improving operational and decision-making processes. In particular, emerging machine intelligence platforms that host pre-trained machine learning models are opening up new opportunities for IoT industries. While those platforms facilitate customers to analyze IoT data and deliver faster and accurate insights, end users and machine learning service providers (MLSPs) have raised concerns regarding security and privacy of IoT data as well as the pre-trained machine learning models for certain applications such as healthcare, smart energy, etc. In this paper, we propose a cloud-assisted, privacy-preserving machine learning classification scheme over encrypted data for IoT devices. Our scheme is based on a three-party model coupled with a two-stage decryption Paillier-based cryptosystem, which allows a cloud server to interact with MLSPs on behalf of the resource-constrained IoT devices in a privacy-preserving manner, and shift load of computation-intensive classification operations from them. The detailed security analysis and the extensive simulations with different key lengths and number of features and classes demonstrate that our scheme can effectively reduce the overhead for IoT devices in machine learning classification applications.

Traditional security controls, such as firewalls, anti-virus and IDS, are ill-equipped to help IT security and response teams keep pace with the rapid evolution of the cyber threat landscape. Cyber Threat Intelligence (CTI) can help remediate this problem by exploiting non-traditional information sources, such as hacker forums and "dark-web" social platforms. Security and response teams can use the collected intelligence to identify emerging threats. Unfortunately, when manual analysis is used to extract CTI from non-traditional sources, it is a time consuming, error-prone and resource intensive process. We address these issues by using a hybrid Machine Learning model that automatically searches through hacker forum posts, identifies the posts that are most relevant to cyber security and then clusters the relevant posts into estimations of the topics that the hackers are discussing. The first (identification) stage uses Support Vector Machines and the second (clustering) stage uses Latent Dirichlet Allocation. We tested our model, using data from an actual hacker forum, to automatically extract information about various threats such as leaked credentials, malicious proxy servers, malware that evades AV detection, etc. The results demonstrate our method is an effective means for quickly extracting relevant and actionable intelligence that can be integrated with traditional security controls to increase their effectiveness.

Despite corporate cyber intrusions attracting all the attention, privacy breaches that we, as ordinary users, should be worried about occur every day without any scrutiny. Smartphones, a household item, have inadvertently become a major enabler of privacy breaches. Smartphone platforms use permission systems to regulate access to sensitive resources. These permission systems, however, lack the ability to understand users’ privacy expectations leaving a significant gap between how permission models behave and how users would want the platform to protect their sensitive data. This dissertation provides an in-depth analysis of how users make privacy decisions in the context of Smartphones and how platforms can accommodate user’s privacy requirements systematically. We first performed a 36-person field study to quantify how often applications access protected resources when users are not expecting it. We found that when the application requesting the permission is running invisibly to the user, they are more likely to deny applications access to protected resources. At least 80% of our participants would have preferred to prevent at least one permission request. To explore the feasibility of predicting user’s privacy decisions based on their past decisions, we performed a longitudinal 131-person field study. Based on the data, we built a classifier to make privacy decisions on the user’s behalf by detecting when the context has changed and inferring privacy preferences based on the user’s past decisions. We showed that our approach can accurately predict users’ privacy decisions 96.8% of the time, which is an 80% reduction in error rate compared to current systems. Based on these findings, we developed a custom Android version with a contextually aware permission model. The new model guards resources based on user’s past decisions under similar contextual circumstances. We performed a 38-person field study to measure the efficiency and usability of the new permission model. Based on exit interviews and 5M data points, we found that the new system is effective in reducing the potential violations by 75%. Despite being significantly more restrictive over the default permission systems, participants did not find the new model to cause any usability issues in terms of application functionality.

User testing is often used to inform the development of user interfaces (UIs). But what if an interface needs to be developed for a system that does not yet exist? In that case, existing datasets can provide valuable input for UI development. We apply a data-driven approach to the development of a privacy-setting interface for Internet-of-Things (IoT) devices. Applying machine learning techniques to an existing dataset of users' sharing preferences in IoT scenarios, we develop a set of "smart" default profiles. Our resulting interface asks users to choose among these profiles, which capture their preferences with an accuracy of 82%—a 14% improvement over a naive default setting and a 12% improvement over a single smart default setting for all users.

With the evolution of network threat, identifying threat from internal is getting more and more difficult. To detect malicious insiders, we move forward a step and propose a novel attribute classification insider threat detection method based on long short term memory recurrent neural networks (LSTM-RNNs). To achieve high detection rate, event aggregator, feature extractor, several attribute classifiers and anomaly calculator are seamlessly integrated into an end-to-end detection framework. Using the CERT insider threat dataset v6.2 and threat detection recall as our performance metric, experimental results validate that the proposed threat detection method greatly outperforms k-Nearest Neighbor, Isolation Forest, Support Vector Machine and Principal Component Analysis based threat detection methods.