Biblio

Face recognition technology is widely employed in a variety of applications, including public security, criminal identification, multimedia data management, and so on. Because of its importance for practical applications and theoretical issues, the facial recognition system has received a lot of attention. Furthermore, numerous strategies have been offered, each of which has shown to be a significant benefit in the field of facial and pattern recognition systems. Face recognition still faces substantial hurdles in unrestricted situations, despite these advancements. Deep learning techniques for facial recognition are presented in this paper for accurate detection and identification of facial images. The primary goal of facial recognition is to recognize and validate facial features. The database consists of 500 color images of people that have been pre-processed and features extracted using Linear Discriminant Analysis. These features are split into 70 percent for training and 30 percent for testing of decision tree classifiers for the computation of face recognition system performance.

Nowadays most of our data is stored on an electronic device. The risk of that device getting infected by Viruses, Malware, Worms, Trojan, Ransomware, or any unwanted invader has increased a lot these days. This is mainly because of easy access to the internet. Viruses and malware have evolved over time so identification of these files has become difficult. Not only by viruses and malware your device can be attacked by a click on forged URLs. Our proposed solution for this problem uses machine learning techniques and signature matching techniques. The main aim of our solution is to identify the malicious programs/URLs and act upon them. The core idea in identifying the malware is selecting the key features from the Portable Executable file headers using these features we trained a random forest model. This RF model will be used for scanning a file and determining if that file is malicious or not. For identification of the virus, we are using the signature matching technique which is used to match the MD5 hash of the file with the virus signature database containing the MD5 hash of the identified viruses and their families. To distinguish between benign and illegitimate URLs there is a logistic regression model used. The regression model uses a tokenizer for feature extraction from the URL that is to be classified. The tokenizer separates all the domains, sub-domains and separates the URLs on every `/'. Then a TfidfVectorizer (Term Frequency - Inverse Document Frequency) is used to convert the text into a weighted value. These values are used to predict if the URL is safe to visit or not. On the integration of all three modules, the final application will provide full system protection against malicious software.

In this paper, we apply machine learning to distributed private data owned by multiple data owners, entities with access to non-overlapping training datasets. We use noisy, differentially-private gradients to minimize the fitness cost of the machine learning model using stochastic gradient descent. We quantify the quality of the trained model, using the fitness cost, as a function of privacy budget and size of the distributed datasets to capture the trade-off between privacy and utility in machine learning. This way, we can predict the outcome of collaboration among privacy-aware data owners prior to executing potentially computationally-expensive machine learning algorithms. Particularly, we show that the difference between the fitness of the trained machine learning model using differentially-private gradient queries and the fitness of the trained machine model in the absence of any privacy concerns is inversely proportional to the size of the training datasets squared and the privacy budget squared. We successfully validate the performance prediction with the actual performance of the proposed privacy-aware learning algorithms, applied to: financial datasets for determining interest rates of loans using regression; and detecting credit card frauds using support vector machines.

A fundamental recurring task in many machine learning applications is the search for the Nearest Neighbor in high dimensional metric spaces. Towards answering queries in large scale problems, state-of-the-art methods employ Approximate Nearest Neighbors (ANN) search, a search that returns the nearest neighbor with high probability, as well as techniques that compress the dataset. Product-Quantization (PQ) based ANN search methods have demonstrated state-of-the-art performance in several problems, including classification, regression and information retrieval. The dataset is encoded into a Cartesian product of multiple low-dimensional codebooks, enabling faster search and higher compression. Being intrinsically parallel, PQ-based ANN search approaches are amendable for hardware acceleration. This paper proposes a novel Hierarchical PQ (HPQ) based ANN search method as well as an FPGA-tailored architecture for its implementation that outperforms current state of the art systems. HPQ gradually refines the search space, reducing the number of data compares and enabling a pipelined search. The mapping of the architecture on a Stratix 10 FPGA device demonstrates over ×250 speedups over current state-of-the-art systems, opening the space for addressing larger datasets and/or improving the query times of current systems.

With the exponential hike in cyber threats, organizations are now striving for better data mining techniques in order to analyze security logs received from their IT infrastructures to ensure effective and automated cyber threat detection. Machine Learning (ML) based analytics for security machine data is the next emerging trend in cyber security, aimed at mining security data to uncover advanced targeted cyber threats actors and minimizing the operational overheads of maintaining static correlation rules. However, selection of optimal machine learning algorithm for security log analytics still remains an impeding factor against the success of data science in cyber security due to the risk of large number of false-positive detections, especially in the case of large-scale or global Security Operations Center (SOC) environments. This fact brings a dire need for an efficient machine learning based cyber threat detection model, capable of minimizing the false detection rates. In this paper, we are proposing optimal machine learning algorithms with their implementation framework based on analytical and empirical evaluations of gathered results, while using various prediction, classification and forecasting algorithms.

Trust in Social Internet of Things has allowed to open new horizons in collaborative networking, particularly by allowing objects to communicate with their service providers, based on their relationships analogy to human world. However, strengthening trust is a challenging task as it involves identifying several influential factors in each domain of social-cyber-physical systems in order to build a reliable system. In this paper, we address the issue of understanding and evaluating honesty that is an important trust metric in trustworthiness evaluation process in social networks. First, we identify and define several trust attributes, which affect directly to the honesty. Then, a subjective computational model is derived based on experiences of objects and opinions from friendly objects with respect to identified attributes. Based on the outputs of this model a final honest level is predicted using regression analysis. Finally, the effectiveness of our model is tested using simulations.

Sharing and working on sensitive data in distributed settings from healthcare to finance is a major challenge due to security and privacy concerns. Secure multiparty computation (SMC) is a viable panacea for this, allowing distributed parties to make computations while the parties learn nothing about their data, but the final result. Although SMC is instrumental in such distributed settings, it does not provide any guarantees not to leak any information about individuals to adversaries. Differential privacy (DP) can be utilized to address this; however, achieving SMC with DP is not a trivial task, either. In this paper, we propose a novel Secure Multiparty Distributed Differentially Private (SM-DDP) protocol to achieve secure and private computations in a multiparty environment. Specifically, with our protocol, we simultaneously achieve SMC and DP in distributed settings focusing on linear regression on horizontally distributed data. That is, parties do not see each others’ data and further, can not infer information about individuals from the final constructed statistical model. Any statistical model function that allows independent calculation of local statistics can be computed through our protocol. The protocol implements homomorphic encryption for SMC and functional mechanism for DP to achieve the desired security and privacy guarantees. In this work, we first introduce the theoretical foundation for the SM-DDP protocol and then evaluate its efficacy and performance on two different datasets. Our results show that one can achieve individual-level privacy through the proposed protocol with distributed DP, which is independently applied by each party in a distributed fashion. Moreover, our results also show that the SM-DDP protocol incurs minimal computational overhead, is scalable, and provides security and privacy guarantees.

The number of wearable and smart devices which are connecting every day in the Internet of Things (IoT) is continuously growing. We have a great opportunity though to improve the quality of life (QoL) standards by adding medical value to these devices. Especially, by exploiting IoT technology, we have the potential to create useful tools which utilize the sensors to provide biometric data. This novel study aims to use a smartwatch, independent from other hardware, to predict the Blood Pressure (BP) drop caused by postural changes. In cases that the drop is due to orthostatic hypotension (OH) can cause dizziness or even faint factors, which increase the risk of fall in the elderly but, as well as, in younger groups of people. A mathematical prediction model is proposed here which can reduce the risk of fall due to OH by sensing heart rate variability (data and drops in systolic BP after standing in a healthy group of 10 subjects. The experimental results justify the efficiency of the model, as it can perform correct prediction in 86.7% of the cases, and are encouraging enough for extending the proposed approach to pathological cases, such as patients with Parkinson's disease, involving large scale experiments.

Online Social Networks have emerged as an interesting area for analysis where each user having a personalized user profile interact and share information with each other. Apart from analyzing the structural characteristics, detection of abnormal and anomalous activities in social networks has become need of the hour. These anomalous activities represent the rare and mischievous activities that take place in the network. Graphical structure of social networks has encouraged the researchers to use various graph metrics to detect the anomalous activities. One such measure that seemed to be highly beneficial to detect the anomalies was brokerage value which helped to detect the anomalies with high accuracy. Also, further application of the measure to different datasets verified the fact that the anomalous behavior detected by the proposed measure was efficient as compared to the already proposed measures in Oddball Algorithm.

One of the major threats against web applications is Cross-Site Scripting (XSS). The final target of XSS attacks is the client running a particular web browser. During this last decade, several competing web browsers (IE, Netscape, Chrome, Firefox) have evolved to support new features. In this paper, we explore whether the evolution of web browsers is done using systematic security regression testing. Beginning with an analysis of their current exposure degree to XSS, we extend the empirical study to a decade of most popular web browser versions. We use XSS attack vectors as unit test cases and we propose a new method supported by a tool to address this XSS vector testing issue. The analysis on a decade releases of most popular web browsers including mobile ones shows an urgent need of XSS regression testing. We advocate the use of a shared security testing benchmark as a good practice and propose a first set of publicly available XSS vectors as a basis to ensure that security is not sacrificed when a new version is delivered.