Biblio

Distant supervision is a widely used data labeling method for relation extraction. While aligning knowledge base with the corpus, distant supervision leads to a mass of wrong labels which are defined as noise. The pattern-based denoising model has achieved great progress in selecting trustable sentences (instances). However, the writing of relation-specific patterns heavily relies on expert’s knowledge and is a high labor intensity work. To solve these problems, we propose a noise reduction framework, NOIR, to iteratively select trustable sentences with a little help of a human. Under the guidance of experts, the iterative process can avoid semantic drift. Besides, NOIR can help experts discover relation-specific tokens that are hard to think of. Experimental results on three real-world datasets show the effectiveness of the proposed method compared with state-of-the-art methods.

Deep learning has undergone tremendous advancements in computer vision studies. The training of deep learning neural networks depends on a considerable amount of ground truth datasets. However, labeling ground truth data is a labor-intensive task, particularly for large-volume video analytics applications such as video surveillance and vehicles detection for autonomous driving. This paper presents a rapid and accurate method for associative searching in big image data obtained from security monitoring systems. We developed a semi-automatic moving object annotation method for improving deep learning models. The proposed method comprises three stages, namely automatic foreground object extraction, object annotation in subsequent video frames, and dataset construction using human-in-the-loop quick selection. Furthermore, the proposed method expedites dataset collection and ground truth annotation processes. In contrast to data augmentation and data generative models, the proposed method produces a large amount of real data, which may facilitate training results and avoid adverse effects engendered by artifactual data. We applied the constructed annotation dataset to train a deep learning you-only-look-once (YOLO) model to perform vehicle detection on street intersection surveillance videos. Experimental results demonstrated that the accurate detection performance was improved from a mean average precision (mAP) of 83.99 to 88.03.

Among the operational shortfalls that hinder law enforcement from achieving greater success in preventing terrorist attacks is the difficulty in dynamically assessing individualized violent extremism risk at scale given the enormous amount of primarily text-based records in disparate databases. In this work, we undertake the critical task of employing natural language processing (NLP) techniques and supervised machine learning models to classify textual data in analyst and investigator notes and reports for radicalization behavioral indicators. This effort to generate structured knowledge will build towards an operational capability to assist analysts in rapidly mining law enforcement and intelligence databases for cues and risk indicators. In the near-term, this effort also enables more rapid coding of biographical radicalization profiles to augment a research database of violent extremists and their exhibited behavioral indicators.

X-ray baggage security screening is a demanding task for aviation and rail transit security; automatic prohibited item detection in X-ray baggage images can help reduce the work of inspectors. However, as many items are placed too close to each other in the baggages, it is difficult to fully trust the detection results of intelligent prohibited item detection algorithms. In this paper, a human-in-the-loop baggage inspection framework is proposed. The proposed framework utilizes the deep-learning-based algorithm for prohibited item detection to find suspicious items in X-ray baggage images, and select manual examination when the detection algorithm cannot determine whether the baggage is dangerous or safe. The advantages of proposed inspection process include: online to capture new sample images for training incrementally prohibited item detection model, and augmented prohibited item detection intelligence with human-computer collaboration. The preliminary experimental results show, human-in-the-loop process by combining cognitive capabilities of human inspector with the intelligent algorithms capabilities, can greatly improve the efficiency of in-baggage security screening.

In contrast to traditional physical servers, a custom-built system utilizing a bare-metal hypervisor virtual server environment provides advantages of both cost savings and flexibility in terms of systems configuration. This system is designed to facilitate hands-on experience for Computer Science students, particularly those specializing in systems administration and computer networking. This multi-purpose and functional system uses an automatic advanced virtual server reservation system (AAVSRsv), written in C++, to schedule and manage virtual servers. The use of such a system could be extended to additional courses focusing on such topics as cloud computing, database systems, information assurance, as well as ethical hacking and system defense. The design can also be replicated to offer training sessions to other information technology professionals.

Online news articles are an important source of information for decisions makers to understand the causal relation of events that happened. However, understanding the causality of an event or between events by traditional machine learning-based techniques from natural language text is a challenging task due to the complexity of the language to be comprehended by the machines. In this study, the Knowledge-oriented convolutional neural network (K-CNN) technique is used to extract the causal relation from online news articles related to the South China Sea (SCS) dispute. The proposed K-CNN model contains a Knowledge-oriented channel that can capture the causal phrases of causal relationships. A Data-oriented channel that captures the position information was added to the K-CNN model in this phase. The online news articles were collected from the national news agency and then the sentences which contain relation such as causal, message-topic, and product-producer were extracted. Then, the extracted sentences were annotated and converted into lower form and base form followed by transformed into the vector by looking up the word embedding table. A word filter that contains causal keywords was generated and a K-CNN model was developed, trained, and tested using the collected data. Finally, different architectures of the K-CNN model were compared to find out the most suitable architecture for this study. From the study, it was found out that the most suitable architecture was the K-CNN model with a Knowledge-oriented channel and a Data-oriented channel with average pooling. This shows that the linguistic clues and the position features can improve the performance in extracting the causal relation from the SCS online news articles. Keywords-component; Convolutional Neural Network, Causal Relation Extraction, South China Sea.

This work presents a system which allows image data and extracted features from a real-world location to be captured and modelled in a Virtual Reality (VR) environment combined with Augmented Situated Visualizations (ASV) overlaid and registered in a virtual environment. Combining these technologies with techniques from Data Science and Artificial Intelligence (AI)(such as image analysis and 3D reconstruction) allows the creation of a setting where remote locations can be modelled and interacted with from anywhere in the world. This Enhanced Photorealistic Immersive (EPI) system is highly adaptable to a wide range of use cases and users as it can be utilized to model and interact with any environment which can be captured as image data (such as training for operation in hazardous environments, accessibility solutions for exploration of historical/tourism locations and collaborative learning environments). A use case example focused on a structural examination of railway tunnels along with a pilot study is presented, which can demonstrate the usefulness of the EPI system.

Since a training system using VR can reproduce an actual training environment, training systems have been studied in commercial fields such as medical care and construction. This immersive experience in a virtual space can have a great effect on learning a second language. In this paper, we propose an immersive learning system that learns phrases used in the customer service industry in the customer service experience. We asked the subjects to experience the system, measured the effects of learning, and evaluated the system. Evaluating the learning effect of phrases used in customer service English on 8 students, all student achieved good learning results. Besides, to evaluate the usability of the system, the VR system was evaluated by performing SSQ to measure VR sickness shows this system doesn't cause virtual sickness, SUS to measure usability shows this system evaluation is higher than average system, and IPQ to measure presence in an immersive space shows this system gives average virtual reality experience.

With the development of deep learning and the progress of natural language processing technology, as well as the continuous disclosure of judicial data such as judicial documents, legal intelligence has gradually become a research hot spot. The crime classification task is an important branch of text classification, which can help people related to the law to improve their work efficiency. However, in the actual research, the sample data is small and the distribution of crime categories is not balanced. To solve these two problems, BERT was used as the encoder to solve the problem of small data volume, and attribute extraction network was added to solve the problem of unbalanced distribution. Finally, the accuracy of 90.35% on small sample data set could be achieved, and F1 value was 67.62, which was close to the best model performance under sufficient data. Finally, a text enhancement method based on back-translation technology is proposed. Different models are used to conduct experiments. Finally, it is found that LSTM model is improved to some extent, but BERT is not improved to some extent.

At present, machine learning (ML) algorithms are essential components in designing the sophisticated intrusion detection system (IDS). They are building-blocks to enhance cyber threat detection and help in classification at host-level and network-level in a short period. The increasing global connectivity and advancements of network technologies have added unprecedented challenges and opportunities to network security. Malicious attacks impose a huge security threat and warrant scalable solutions to thwart large-scale attacks. These activities encourage researchers to address these imminent threats by analyzing a large volume of the dataset to tackle all possible ranges of attack. In this proposed method, we calculated the fitness value of each feature from the population by using a genetic algorithm (GA) and selected them according to the fitness value. The fitness values are presented in hierarchical order to show the effectiveness of problem decomposition. We implemented Support Vector Machine (SVM) to verify the consistency of the system outcome. The well-known NSL-knowledge discovery in databases (KDD) was used to measure the performance of the system. From the experiments, we achieved a notable classification accuracies using a SVM of the current state of the art intrusion detection.

The extraordinary number of alerts generated by network intrusion detection systems (NIDS) can desensitize security analysts tasked with incident response. Security information and event management systems (SIEMs) perform some rudimentary automation but cannot replicate the decision-making process of a skilled analyst. Machine learning and artificial intelligence (AI) can detect patterns in data with appropriate training. In practice, the majority of the alert data comprises false alerts, and true alerts form only a small proportion. Consequently, a naive engine that classifies all security alerts into the majority class can yield a superficial high accuracy close to 100%. Without any correction for the class imbalance, the false alerts will dominate algorithmic predictions resulting in poor generalization performance. We propose a machine-learning approach to address the class imbalance problem in multi-appliance security alert data and automate the security alert analysis process performed in security operations centers (SOCs). We first used the neighborhood cleaning rule (NCR) to identify and remove ambiguous, noisy, and redundant false alerts. Then, we applied the support vector machine synthetic minority oversampling technique (SVMSMOTE) to generate synthetic training true alerts. Finally, we fit and evaluated the decision tree and random forest classifiers. In the experiments, using alert data from eight security appliances, we demonstrated that the proposed method can significantly reduce the need for manual auditing, decreasing the number of uninspected alerts and achieving a performance of 99.524% in recall.

In new technological world pervasive computing plays the important role in data computing and communication. The pervasive computing provides the mobile environment for decentralized computational services at anywhere, anytime at any context and location. Pervasive computing is flexible and makes portable devices and computing surrounded us as part of our daily life. Devices like Laptop, Smartphones, PDAs, and any other portable devices can constitute the pervasive environment. These devices in pervasive environments are worldwide and can receive various communications including audio visual services. The users and the system in this pervasive environment face the challenges of user trust, data privacy and user and device node identity. To give the feasible determination for these challenges. This paper aims to propose a dynamic learning in pervasive computing environment refer the challenges proposed efficient security model (ESM) for trustworthy and untrustworthy attackers. ESM model also compared with existing generic models; it also provides better accuracy rate than existing models.

With the rapid development of e-commerce and the increasing popularity of credit cards, online transactions have become increasingly smooth and convenient. However, many online transactions suffer from credit card fraud, resulting in huge losses every year. Many financial organizations and e-commerce companies are devoted to developing advanced fraud detection algorithms. This paper presents an approach to detect fraud transactions using the IEEE-CIS Fraud Detection dataset provided by Kaggle. Our stacked model is based on Gradient Boosting, LightGBM, CatBoost, and Random Forest. Besides, implementing StackNet improves the classification accuracy significantly and provides expandability to the network architecture. Our final model achieved an AUC of 0.9578 for the training set and 0.9325 for the validation set, demonstrating excellent performance in classifying different transaction types.

With increased awareness comes unprecedented expectations. We live in a digital, cloud era wherein the underlying information architectures are expected to be elastic, secure, resilient, and handle petabyte scaling. The expectation of epic proportions from the next generation of the data frameworks is to not only do all of the above but also build it on a foundation of trust and explainability across multi-organization business networks. From cloud providers to automobile industries or even vaccine manufacturers, components are often sourced by a complex, not full digitized thread of disjoint suppliers. Building Machine Learning and AI-based order fulfillment and predictive models, remediating issues, is a challenge for multi-organization supply chain automation. We posit that Federated Learning in conjunction with blockchain and smart contracts are technologies primed to tackle data privacy and centralization challenges. In this paper, motivated by challenges in the industry, we propose a decentralized distributed system in conjunction with a recommendation system model (Matrix Factorization) that is trained using Federated Learning on an Ethereum blockchain network. We leverage smart contracts that allow decentralized serverless aggregation to update local-ized items vectors. Furthermore, we utilize Homomorphic Encryption (HE) to allow sharing the encrypted gradients over the network while maintaining their privacy. Based on our results, we argue that training a model over a serverless Blockchain network using smart contracts will provide the same accuracy as in a centralized model while maintaining our serverless model privacy and reducing the overhead communication to a central server. Finally, we assert such a system that provides transparency, audit-ready and deep insights into supply chain operations for enterprise cloud customers resulting in cost savings and higher Quality of Service (QoS).

Edge nodes are crucial for detection against multitudes of cyber attacks on Internet-of-Things endpoints and is set to become part of a multi-billion industry. The resource constraints in this novel network infrastructure tier constricts the deployment of existing Network Intrusion Detection System with Deep Learning models (DLM). We address this issue by developing a novel light, fast and accurate `Edge-Detect' model, which detects Distributed Denial of Service attack on edge nodes using DLM techniques. Our model can work within resource restrictions i.e. low power, memory and processing capabilities, to produce accurate results at a meaningful pace. It is built by creating layers of Long Short-Term Memory or Gated Recurrent Unit based cells, which are known for their excellent representation of sequential data. We designed a practical data science pipeline with Recurring Neural Network to learn from the network packet behavior in order to identify whether it is normal or attack-oriented. The model evaluation is from deployment on actual edge node represented by Raspberry Pi using current cybersecurity dataset (UNSW2015). Our results demonstrate that in comparison to conventional DLM techniques, our model maintains a high testing accuracy of 99% even with lower resource utilization in terms of cpu and memory. In addition, it is nearly 3 times smaller in size than the state-of-art model and yet requires a much lower testing time.

Deep learning has made remarkable achievements in various domains. Active learning, which aims to reduce the budget for training a machine-learning model, is especially useful for the Deep learning tasks with the demand of a large number of labeled samples. Unfortunately, our empirical study finds that many of the active learning heuristics are not effective when applied to Deep learning models in batch settings. To tackle these limitations, we propose a density weighted diversity based query strategy (DWDS), which makes use of the geometry of the samples. Within a limited labeling budget, DWDS enhances model performance by querying labels for the new training samples with the maximum informativeness and representativeness. Furthermore, we propose a beam-search based method to obtain a good approximation to the optimum of such samples. Our experiments show that DWDS outperforms existing algorithms in Deep learning tasks.

Keystroke dynamics is a behavioural biometric form of authentication based on the inherent typing behaviour of an individual. While this technique is gaining traction, protecting the privacy of the users is of utmost importance. Fully Homomorphic Encryption is a technique that allows performing computation on encrypted data, which enables processing of sensitive data in an untrusted environment. FHE is also known to be “future-proof” since it is a lattice-based cryptosystem that is regarded as quantum-safe. It has seen significant performance improvements over the years with substantially increased developer-friendly tools. We propose a neural network for keystroke analysis trained using differential privacy to speed up training while preserving privacy and predicting on encrypted data using FHE to keep the users' privacy intact while offering sufficient usability.

With the development of network, network security has become a topic of increasing concern. Recent years, machine learning technology has become an effective means of network intrusion detection. However, machine learning technology requires a large amount of data for training, and training data often contains privacy information, which brings a great risk of privacy leakage. At present, there are few researches on data privacy protection in the field of intrusion detection. Regarding the issue of privacy and security, we combine differential privacy and machine learning algorithms, including One-class Support Vector Machine (OCSVM) and Local Outlier Factor(LOF), to propose an hybrid intrusion detection system (IDS) with privacy protection. We add Laplacian noise to the original network intrusion detection data set to get differential privacy data sets with different privacy budgets, and proposed a hybrid IDS model based on machine learning to verify their utility. Experiments show that while protecting data privacy, the hybrid IDS can achieve detection accuracy comparable to traditional machine learning algorithms.

Federated learning is a distributed learning technique where machine learning models are trained on client devices in which the local training data resides. The training is coordinated via a central server which is, typically, controlled by the intended owner of the resulting model. By avoiding the need to transport the training data to the central server, federated learning improves privacy and efficiency. But it raises the risk of model theft by clients because the resulting model is available on every client device. Even if the application software used for local training may attempt to prevent direct access to the model, a malicious client may bypass any such restrictions by reverse engineering the application software. Watermarking is a well-known deterrence method against model theft by providing the means for model owners to demonstrate ownership of their models. Several recent deep neural network (DNN) watermarking techniques use backdooring: training the models with additional mislabeled data. Backdooring requires full access to the training data and control of the training process. This is feasible when a single party trains the model in a centralized manner, but not in a federated learning setting where the training process and training data are distributed among several client devices. In this paper, we present WAFFLE, the first approach to watermark DNN models trained using federated learning. It introduces a retraining step at the server after each aggregation of local models into the global model. We show that WAFFLE efficiently embeds a resilient watermark into models incurring only negligible degradation in test accuracy (-0.17%), and does not require access to training data. We also introduce a novel technique to generate the backdoor used as a watermark. It outperforms prior techniques, imposing no communication, and low computational (+3.2%) overhead$^\textrm1$$^\textrm1$\$The research report version of this paper is also available in https://arxiv.org/abs/2008.07298, and the code for reproducing our work can be found at https://github.com/ssg-research/WAFFLE.

Asymmetric warfare and anti-terrorist war have become a new style of military struggle in the new century, which will inevitably have an important impact on the military economy of various countries and catalyze the innovation climax of military logistics theory and practice. The war in the information age is the confrontation between systems, and “comprehensive integration” is not only the idea of information war ability construction, but also the idea of deterrence ability construction in the information age. Looking at the local wars under the conditions of modern informationization, it is not difficult to see that the status and role of light weapons and equipment have not decreased, on the contrary, higher demands have been put forward for their combat performance. From a forward-looking perspective, based on our army's preparation and logistics support for future asymmetric operations and anti-terrorist military struggle, this strategic issue is discussed in depth.

This paper aims to address the problem of Deepfake Detection using a Two-Stream Capsule Network. First we review methods used to create Deepfake content, as well as methods proposed in the literature to detect such Deepfake content. We then propose a novel architecture to detect Deepfakes, which consists of a two-stream Capsule network running in parallel that takes in both RGB images/frames as well as Error Level Analysis images. Results show that the proposed approach exhibits the detection accuracy of 73.39 % and 57.45 % for the Deepfake Detection Challenge (DFDC) and the Celeb-DF datasets respectively. These results are, however, from a preliminary implementation of the proposed approach. As part of future work, population-based optimization techniques such as Particle Swarm Optimization (PSO) will be used to tune the hyper parameters for better performance.

Nowadays, developing an intelligent system for segmenting the moving object from the background is essential task for video surveillance applications. Recently, a deep learning segmentation algorithm composed of encoder CNN, a Feature Pooling Module and a decoder CNN called FgSegNET\_S has been proposed. It is capable to train the model using few training examples. FgSegNET\_S is relying only on the spatial information while it is fundamental to include temporal information to distinguish if an object is moving or not. In this paper, an improved version known as (T\_FgSegNET\_S) is proposed by using the subtracted images from the initial background as input. The proposed approach is trained and evaluated using two publicly available infrared datasets: remote scene infrared videos captured by medium-wave infrared (MWIR) sensors and the Grayscale Thermal Foreground Detection (GTFD) dataset. The performance of network is evaluated using precision, recall, and F-measure metrics. The experiments show improved results, especially when compared to other state-of-the-art methods.

This paper designs a data anomaly detection method for power grid data centers. The method uses cloud computing architecture to realize the storage and calculation of large amounts of data from power grid data centers. After that, the STL decomposition method is used to decompose the grid data, and then the decomposed residual data is used for anomaly analysis to complete the detection of abnormal data in the grid data. Finally, the feasibility of the method is verified through experiments.

Covert communications enable a transmitter to send information reliably in the presence of an adversary, who looks to detect whether the transmission took place or not. We consider covert communications over quasi-static block fading channels, where users suffer from channel uncertainty. We investigate the adversary Willie's optimal detection performance in two extreme cases, i.e., the case of perfect channel state information (CSI) and the case of channel distribution information (CDI) only. It is shown that in the large detection error regime, Willie's detection performances of these two cases are essentially indistinguishable, which implies that the quality of CSI does not help Willie in improving his detection performance. This result enables us to study the covert transmission design without the need to factor in the exact amount of channel uncertainty at Willie. We then obtain the optimal and suboptimal closed-form solution to the covert transmission design. Our result reveals fundamental difference in the design between the case of quasi-static fading channel and the previously studied case of non-fading AWGN channel.

Web application firewalls (WAF) are the last line of defense in protecting web applications from application layer security threats like SQL injection and cross-site scripting. Currently, most evasion techniques from WAFs are still developed manually. In this work, we propose a solution, which automatically scans the WAFs to find payloads through which the WAFs can be bypassed. Our solution finds out rules defects, which can be further used in rule tuning for rule-based WAFs. Also, it can enrich the machine learning-based dataset for retraining. To this purpose, we provide a framework based on reinforcement learning with an environment compatible with OpenAI gym toolset standards, employed for training agents to implement WAF evasion tasks. The framework acts as an adversary and exploits a set of mutation operators to mutate the malicious payload syntactically without affecting the original semantics. We use Q-learning and proximal policy optimization algorithms with the deep neural network. Our solution is successful in evading signature-based and machine learning-based WAFs.