World Academy of Science, Engineering and Technology

Authors: Ashly Joseph, Jithu Paulose

Abstract:

The increased usage of cloud computing has revolutionized the IT landscape, but it has also raised new security concerns. DevSecOps emerged as a way for tackling these difficulties by integrating security into the software development process. However, the rising complexity and sophistication of cyber threats need more advanced solutions. This paper looks into the usage of artificial intelligence (AI) techniques in the DevSecOps framework to increase cloud security. This study uses quantitative and qualitative techniques to assess the usefulness of AI approaches such as machine learning, natural language processing, and deep learning in reducing security issues. This paper thoroughly examines the symbiotic relationship between AI and DevSecOps, concentrating on how AI may be seamlessly integrated into the continuous integration and continuous delivery (CI/CD) pipeline, automated security testing, and real-time monitoring methods. The findings emphasize AI's huge potential to improve threat detection, risk assessment, and incident response skills. Furthermore, the paper examines the implications and challenges of using AI in DevSecOps workflows, considering factors like as scalability, interpretability, and adaptability. This paper adds to a better understanding of AI's revolutionary role in cloud security and provides valuable insights for practitioners and scholars in the field.

Keywords: Cloud Security, DevSecOps, Artificial Intelligence, AI, Machine Learning, Natural Language Processing, NLP, cybersecurity, AI-driven Security.

Authors: H. AbdelRahman, S. Rostom, Y. Lotfy, S. Salah Eldeen, R. Yassein, N. Awny

Abstract:

People are growing more concerned with their appearance in today's society. But they are terrified of what they will look like after a plastic surgery. People's mental health suffers when they have accidents, burns, or genetic issues that cause them to cleave certain body parts, which makes them feel uncomfortable and unappreciated. The method provides an innovative deep learning-based technique for image inpainting that analyzes different picture structures and fixes damaged images. This study proposes a model based on the Stable Diffusion Inpainting method for in-painting medical images. One significant advancement made possible by deep neural networks is image inpainting, which is the process of reconstructing damaged and missing portions of an image. The patient can see the outcome more easily since the system uses the user's input of an image to identify a problem. It then modifies the image and outputs a fixed image.

Keywords: Generative Adversarial Network, GAN, Large Mask Inpainting, LAMA, Stable Diffusion Inpainting.

Authors: Luisa Mesquita Oliveira Ribeiro, Alexei Manso Correa Machado

Abstract:

Handwritten signature is a unique form for recognizing an individual, used to discern documents, carry out investigations in the criminal, legal, banking areas and other applications. Signature verification is based on large amounts of biometric data, as they are simple and easy to acquire, among other characteristics. Given this scenario, signature forgery is a worldwide recurring problem and fast and precise techniques are needed to prevent crimes of this nature from occurring. This article carried out a study on the efficiency of the Capsule Network in analyzing and recognizing signatures. The chosen architecture achieved an accuracy of 98.11% and 80.15% for the CEDAR and GPDS databases, respectively.

Keywords: Biometrics, deep learning, handwriting, signature forgery.

Authors: Cheng Zeng, George Michailidis, Hitoshi Iyatomi, Leo L Duan

Abstract:

The conditional density characterizes the distribution of a response variable y given other predictor x, and plays a key role in many statistical tasks, including classification and outlier detection. Although there has been abundant work on the problem of Conditional Density Estimation (CDE) for a low-dimensional response in the presence of a high-dimensional predictor, little work has been done for a high-dimensional response such as images. The promising performance of normalizing flow (NF) neural networks in unconditional density estimation acts a motivating starting point. In this work, we extend NF neural networks when external x is present. Specifically, they use the NF to parameterize a one-to-one transform between a high-dimensional y and a latent z that comprises two components [zP , zN]. The zP component is a low-dimensional subvector obtained from the posterior distribution of an elementary predictive model for x, such as logistic/linear regression. The zN component is a high-dimensional independent Gaussian vector, which explains the variations in y not or less related to x. Unlike existing CDE methods, the proposed approach, coined Augmented Posterior CDE (AP-CDE), only requires a simple modification on the common normalizing flow framework, while significantly improving the interpretation of the latent component, since zP represents a supervised dimension reduction. In image analytics applications, AP-CDE shows good separation of x-related variations due to factors such as lighting condition and subject id, from the other random variations. Further, the experiments show that an unconditional NF neural network, based on an unsupervised model of z, such as Gaussian mixture, fails to generate interpretable results.

Keywords: Conditional density estimation, image generation, normalizing flow, supervised dimension reduction.

Authors: Yi Yang, Jiasong Sun

Abstract:

Almost every aspect of our daily lives is now intertwined with some degree of Human-Computer Interaction (HCI). HCI courses draw on knowledge from disciplines as diverse as computer science, psychology, design principles, anthropology and more. The HCI courses in the Department of Electronics at Tsinghua University, known as the Media and Cognition course, is constantly updated to reflect the most advanced technological advances, such as virtual reality, augmented reality and artificial intelligence-based interaction. For more than a decade, this course has used an interest-based approach to teaching, in which students proactively propose some research-based questions and collaborate with teachers, using course knowledge to explore potential solutions. Semantic communication plays a key role in facilitating understanding and interaction between users and computer systems, ultimately enhancing system usability and user experience. The advancements in AI-generated technology, which has gained significant attention from both academia and industry in recent years, are exemplified by language models like GPT-3 that generate human-like dialogues from given prompts. The latest version of the HCI course practices a semantic communication platform based on AI-generated techniques. We explored a student-centered model and proposed an interest-based teaching method. Students are no longer just recipients of knowledge, but become active participants in the learning process driven by personal interests, thereby encouraging students to take responsibility for their own education. One of the latest results of this teaching approach in the course "Media and Cognition" is a student proposal to develop a semantic communication platform rooted in artificial intelligence generative technologies. The platform solves a key challenge in communications technology: the ability to preserve visual signals. The interest-based approach emphasizes personal curiosity and active participation, and the proposal of an artificial intelligence-generated semantic communication platform is an example and successful result of how students can exert greater creativity when they have the power to control their own learning.

Keywords: Human-computer interaction, media and cognition course, semantic communication, retain ability, prompts.

Authors: Andrew John Zeller, Francis Pouatcha

Abstract:

Working according to the DevOps principle has gained in popularity over the past decade. While its extension DevSecOps started to include elements of cybersecurity, most real-life projects do not focus risk and security until the later phases of a project as teams are often more familiar with engineering and infrastructure services. To help bridge the gap between security and engineering, this paper will take six building blocks of cybersecurity and apply them to the DevOps approach. After giving a brief overview of the stages in the DevOps lifecycle, the main part discusses to what extent six cybersecurity blocks can be utilized in various stages of the lifecycle. The paper concludes with an outlook on how to stay up to date in the dynamic world of cybersecurity.

Keywords: Information security, data security, cybersecurity, DevOps, IT management.

Authors: M. Sitoe, O. Zacarias

Abstract:

University education is influenced by several factors that range from the adoption of strategies to strengthen the whole process to the academic performance improvement of the students themselves. This work uses data mining techniques to develop a predictive model to identify students with a tendency to evasion and retention. To this end, a database of real students’ data from the Department of University Admission (DAU) and the Department of Mathematics and Informatics (DMI) was used. The data comprised 388 undergraduate students admitted in the years 2009 to 2014. The Weka tool was used for model building, using three different techniques, namely: K-nearest neighbor, random forest, and logistic regression. To allow for training on multiple train-test splits, a cross-validation approach was employed with a varying number of folds. To reduce bias variance and improve the performance of the models, ensemble methods of Bagging and Stacking were used. After comparing the results obtained by the three classifiers, Logistic Regression using Bagging with seven folds obtained the best performance, showing results above 90% in all evaluated metrics: accuracy, rate of true positives, and precision. Retention is the most common tendency.

Keywords: Evasion and retention, cross validation, bagging, stacking.

Authors: Amir Tosson, Mohammad Reza, Christian Gutt

Abstract:

Metadata is one of the most important aspects for advancing data management practices within all research communities. Definitions and schemes of metadata are inter alia of particular significance in the domain of neutron and photon scattering experiments covering a broad area of different scientific disciplines. The demand of describing continuously evolving highly non-standardized experiments, including the resulting processed and published data, constitutes a considerable challenge for a static definition of metadata. Here, we present the concept of dynamic metadata for the neutron and photon scientific community, which enriches a static set of defined basic metadata. We explore the idea of dynamic metadata with the help of the use case of X-ray Photon Correlation Spectroscopy (XPCS), which is a synchrotron-based scattering technique that allows the investigation of nanoscale dynamic processes. It serves here as a demonstrator of how dynamic metadata can improve data acquisition, sharing, and analysis workflows. Our approach enables researchers to tailor metadata definitions dynamically and adapt them to the evolving demands of describing data and results from a diverse set of experiments. We demonstrate that dynamic metadata standards yield advantages that enhance data reproducibility, interoperability, and the dissemination of knowledge.

Keywords: Big data, metadata, schemas, XPCS, X-ray Photon Correlation Spectroscopy.

Authors: C. Ardil

Abstract:

Uncertainty set theory is a generalization of fuzzy set theory and intuitionistic fuzzy set theory. It serves as an effective tool for dealing with inconsistent, imprecise, and vague information. The technique for order preference by similarity to ideal solution (TOPSIS) method is a multiple-attribute method used to identify solutions from a finite set of alternatives. It simultaneously minimizes the distance from an ideal point and maximizes the distance from a nadir point. In this paper, an extension of the TOPSIS method for multiple attribute group decision-making (MAGDM) based on uncertainty sets is presented. In uncertainty decision analysis, decision-makers express information about attribute values and weights using uncertainty numbers to select the best stealth combat aircraft.

Keywords: Uncertainty set, stealth combat aircraft selection multiple criteria decision-making analysis, MCDM, uncertainty decision analysis, TOPSIS

Authors: Stephen Akuma, Isaac Terngu Adom, Evelyn Doofan Akuma

Abstract:

Generation after generation of humans has experienced conflicts leading to needless deaths. Usually, it begins as a minor argument that occasionally escalates into a full-fledged conflict. There has been a lingering crisis in the Lake Chad Basin (LCB) of Africa for over a decade leading to bloodshed that has claimed thousands of lives. The terrorist group, Boko Haram has claimed responsibility for these deaths. Efforts have been made by the governments in the LCB region to end the crisis through kinetic approaches, but the conflict persists. In this work, we explored non-kinetic methods used by social scientists in resolving conflicts, with a focus on computational approaches due to the increasing processing power of the computer. Firstly, we reviewed the innovative computational methods available for researchers working on conflict, violence, and peace. Secondly, we described how an Agent-Based Model (ABM) can be linked with a Geographic Information System (GIS) to model the LCB. Finally, this research discusses the challenges and opportunities in constructing a Geographic Information System linked Agent-Based Model of the LCB region.

Keywords: Agent-based modelling, conflict, Geographical Information Systems, Lake Chad Basin, simulation.

Authors: Ashly Joseph

Abstract:

Cloud computing has become an essential component of enterprises and organizations globally in the current era of digital technology. The cloud has a multitude of advantages, including scalability, flexibility, and cost-effectiveness, rendering it an appealing choice for data storage and processing. The increasing storage of sensitive information in cloud environments has raised significant concerns over the security of such systems. The frequency of cyber threats and attacks specifically aimed at cloud infrastructure has been increasing, presenting substantial dangers to the data, reputation, and financial stability of enterprises. Conventional security methods can become inadequate when confronted with ever intricate and dynamic threats. Artificial Intelligence (AI) technologies possess the capacity to significantly transform cloud security through their ability to promptly identify and thwart assaults, adjust to emerging risks, and offer intelligent perspectives for proactive security actions. The objective of this research study is to investigate the utilization of AI technologies in augmenting the security measures within cloud computing systems. This paper aims to offer significant insights and recommendations for businesses seeking to protect their cloud-based assets by analyzing the present state of cloud security, the capabilities of AI, and the possible advantages and obstacles associated with using AI into cloud security policies.

Keywords: Machine Learning, Natural Learning Processing, Denial-of-Service attacks, Sentiment Analysis, Cloud computing.

Authors: Aseela Alhinaai, Suliman Abdullah, Adil Albusaidi

Abstract:

Knowledge has been considered an important asset in private and public organizations. It is utilized in the libraries sector to run different operations of technical services and administrative works. This study aims to identify the impact of the knowledge-sharing factors (technology, collaboration, management support) to improve decision-making at Sultan Qaboos University Libraries. This study conducted a quantitative method using a questionnaire instrument to measure the impact of technology, collaboration, and management support on knowledge sharing that lead to improved decision-making. The study population is the Sultan Qaboos University (SQU) libraries (Main Library, Medical Library, College of Economic and Political Science Library, and Art Library). The results showed that management support, collaboration, and technology use have a positive impact on the knowledge-sharing process, and knowledge sharing positively affects decision making process.

Keywords: Knowledge sharing, decision making, information technology, management support, corroboration, Sultan Qaboos University.

Authors: Kang Zijian, Zhang Tingyu, Burra Venkata Durga Kumar

Abstract:

This article investigates the challenges in memory migration during the live migration of virtual machines. We found three challenges probably existing in pre-copy technology. One of the main challenges is the challenge of downtime migration. Decreasing the downtime could promise the normal work for a virtual machine. Although pre-copy technology is greatly decreasing the downtime, we still need to shut down the machine in order to finish the last round of data transfer. This paper provides an optimization scheme for the problems existing in pro-copy technology, mainly the optimization of the dirty page migration mechanism. The typical pre-copy technology copies n-1th’s dirty pages in nth turn. However, our idea is to create a double iteration method to solve this problem.

Keywords: Virtual machine, pre-copy technology, memory migration process, downtime, dirty pages migration method.

Authors: Alireza Vafaei Sadr, Vida Abedi, Jiang Li, Ramin Zand

Abstract:

Missing data is a common challenge in medical research and can lead to biased or incomplete results. When the data bias leaks into models, it further exacerbates health disparities; biased algorithms can lead to misclassification and reduced resource allocation and monitoring as part of prevention strategies for certain minorities and vulnerable segments of patient populations, which in turn further reduce data footprint from the same population – thus, a vicious cycle. This study compares the performance of six imputation techniques grouped into Linear and Non-Linear models, on two different real-world electronic health records (EHRs) datasets, representing 17864 patient records. The mean absolute percentage error (MAPE) and root mean squared error (RMSE) are used as performance metrics, and the results show that the Linear models outperformed the Non-Linear models in terms of both metrics. These results suggest that sometimes Linear models might be an optimal choice for imputation in laboratory variables in terms of imputation efficiency and uncertainty of predicted values.

Keywords: EHR, Machine Learning, imputation, laboratory variables, algorithmic bias.

Authors: Sarah Beaver, Renee Bryce

Abstract:

Audiograms detect hearing impairment, but missing values pose problems. This work explores imputations in an attempt to improve accuracy. This work implements Linear Regression, Lasso, Linear Support Vector Regression, Bayesian Ridge, K Nearest Neighbors (KNN), and Random Forest machine learning techniques to impute audiogram frequencies ranging from 125 Hz to 8000 Hz. The data contain patients who had or were candidates for cochlear implants. Accuracy is compared across two different Nested Cross-Validation k values. Over 4000 audiograms were used from 800 unique patients. Additionally, training on data combines and compares left and right ear audiograms versus single ear side audiograms. The accuracy achieved using Root Mean Square Error (RMSE) values for the best models for Random Forest ranges from 4.74 to 6.37. The R2 values for the best models for Random Forest ranges from .91 to .96. The accuracy achieved using RMSE values for the best models for KNN ranges from 5.00 to 7.72. The R2 values for the best models for KNN ranges from .89 to .95. The best imputation models received R2 between .89 to .96 and RMSE values less than 8dB. We also show that the accuracy of classification predictive models performed better with our imputation models versus constant imputations by a two percent increase.

Keywords: Machine Learning, audiograms, data imputations, single imputations.

Authors: Yunyong Guo, Sudhakar Ganti, Bryan Guo

Abstract:

The rapid growth of telehealth Internet of Things (IoT) devices has raised concerns about energy consumption and efficient data processing. This paper presents an energy-efficient model that integrates telehealth IoT devices with a fog and cloud computing-based platform, offering a sustainable and robust solution to overcome these challenges. Our model employs fog computing as a localized data processing layer while leveraging cloud computing for resource-intensive tasks, significantly reducing energy consumption. We incorporate adaptive energy-saving strategies. Simulation analysis validates our approach's effectiveness in enhancing energy efficiency for telehealth IoT systems integrated with localized fog nodes and both private and public cloud infrastructures. Future research will focus on further optimization of the energy-saving model, exploring additional functional enhancements, and assessing its broader applicability in other healthcare and industry sectors.

Keywords: Energy-efficient, fog computing, IoT, telehealth.

Authors: Papathanasiou Anastasios, Liontos George, Liagkou Vasiliki, Glavas Euripides

Abstract:

The purpose of this paper is to describe the growing problem of various cyber fraud schemes that exist on the internet and are currently among the most prevalent. The main focus of this paper is to provide a detailed description of the modus operandi, tools, and techniques utilized in four basic typologies of cyber frauds: Business Email Compromise (BEC) attacks, investment fraud, romance scams, and online sales fraud. The paper aims to shed light on the methods employed by cybercriminals in perpetrating these types of fraud, as well as the strategies they use to deceive and victimize individuals and businesses on the internet. Furthermore, this study outlines defense strategies intended to tackle the issue head-on, with a particular emphasis on the crucial role played by European legislation. European legislation has proactively adapted to the evolving landscape of cyber fraud, striving to enhance cybersecurity awareness, bolster user education, and implement advanced technical controls to mitigate associated risks. The paper evaluates the advantages and innovations brought about by the European legislation while also acknowledging potential flaws that cybercriminals might exploit. As a result, recommendations for refining the legislation are offered in this study in order to better address this pressing issue.

Keywords: Business email compromise, cybercrime, European legislation, investment fraud, Network and Information Security, online sales fraud, romance scams.

Authors: Yong Zhao, Jian He, Cheng Zhang

Abstract:

Cardiovascular disease resulting from hypertension poses a significant threat to human health, and early detection of hypertension can potentially save numerous lives. Traditional methods for detecting hypertension require specialized equipment and are often incapable of capturing continuous blood pressure fluctuations. To address this issue, this study starts by analyzing the principle of heart rate variability (HRV) and introduces the utilization of sliding window and power spectral density (PSD) techniques to analyze both temporal and frequency domain features of HRV. Subsequently, a hypertension prediction network that relies on HRV is proposed, combining Resnet, attention mechanisms, and a multi-layer perceptron. The network leverages a modified ResNet18 to extract frequency domain features, while employing an attention mechanism to integrate temporal domain features, thus enabling auxiliary hypertension prediction through the multi-layer perceptron. The proposed network is trained and tested using the publicly available SHAREE dataset from PhysioNet. The results demonstrate that the network achieves a high prediction accuracy of 92.06% for hypertension, surpassing traditional models such as K Near Neighbor (KNN), Bayes, Logistic regression, and traditional Convolutional Neural Network (CNN).

Keywords: Feature extraction, heart rate variability, hypertension, residual networks.

Authors: Manouchehr Amiri

Abstract:

This article presents a general notion of physical bit information that is compatible with the basics of quantum mechanics and incorporates the Shannon entropy as a special case. This notion of physical information leads to the Binary Data Matrix model (BDM), which predicts the basic results of quantum mechanics, general relativity, and black hole thermodynamics. The compatibility of the model with holographic, information conservation, and Landauer’s principle is investigated. After deriving the “Bit Information principle” as a consequence of BDM, the fundamental equations of Planck, De Broglie, Bekenstein, and mass-energy equivalence are derived.

Keywords: Physical theory of information, binary data matrix model, Shannon information theory, bit information principle.

Authors: Sharifah Mousli, Sona Taheri, Jiayuan He

Abstract:

Autism spectrum disorder (ASD) is a complex developmental condition involving persistent difficulties with social communication, restricted interests, and repetitive behavior. The challenges associated with ASD can interfere with an affected individual’s ability to function in social, academic, and employment settings. Although there is no effective medication known to treat ASD, to our best knowledge, early intervention can significantly improve an affected individual’s overall development. Hence, an accurate diagnosis of ASD at an early phase is essential. The use of machine learning approaches improves and speeds up the diagnosis of ASD. In this paper, we focus on the application of unsupervised clustering methods in ASD, as a large volume of ASD data generated through hospitals, therapy centers, and mobile applications has no pre-existing labels. We conduct a comparative analysis using seven clustering approaches, such as K-means, agglomerative hierarchical, model-based, fuzzy-C-means, affinity propagation, self organizing maps, linear vector quantisation – as well as the recently developed optimization-based clustering (COMSEP-Clust) approach. We evaluate the performances of the clustering methods extensively on real-world ASD datasets encompassing different age groups: toddlers, children, adolescents, and adults. Our experimental results suggest that the COMSEP-Clust approach outperforms the other seven methods in recognizing ASD with well-separated clusters.

Keywords: Autism spectrum disorder, clustering, optimization, unsupervised machine learning.

Authors: Sarah Vuningoma, Maria Rosa Lorini, Wallace Chigona

Abstract:

The potential of mobile phones is evident in their ability to address isolation and loneliness, support the improvement of interpersonal relations, and contribute to the facilitation of assimilation processes. Mobile phones can play a role in facilitating the integration of refugees into a new environment. This study aims to evaluate the impact of mobile phone use on helping refugees navigate the challenges posed by cultural differences in the host country. Semi-structured interviews were employed to collect data for the study, involving a sample size of 27 participants. Participants in the study were refugees based in South Africa, and thematic analysis was the chosen method for data analysis. The research highlights the numerous challenges faced by refugees in their host nation, including a lack of local cultural skills, the separation of family and friends from their countries of origin, hurdles in acquiring legal documentation, and the complexities of assimilating into the unfamiliar community. The use of mobile phones by refugees comes with several advantages, such as the advancement of language and cultural understanding, seamless integration into the host country, streamlined communication, and the exploration of diverse opportunities. Concurrently, mobile phones allow refugees in South Africa to manage the impact of culture shock.

Keywords: Mobile phones, culture shock, refugees, South Africa.

Authors: Abdul Basit Kiani, Maryam Kiani

Abstract:

The art of web development in new technologies is a dynamic journey, shaped by the constant evolution of tools and platforms. With the emergence of JavaScript frameworks and APIs, web developers are empowered to craft web applications that are not only robust but also highly interactive. The aim is to provide an overview of the developments in the field. The integration of artificial intelligence (AI) and machine learning (ML) has opened new horizons in web development. Chatbots, intelligent recommendation systems, and personalization algorithms have become integral components of modern websites. These AI-powered features enhance user engagement, provide personalized experiences, and streamline customer support processes, revolutionizing the way businesses interact with their audiences. Lastly, the emphasis on web security and privacy has been a pivotal area of progress. With the increasing incidents of cyber threats, web developers have implemented robust security measures to safeguard user data and ensure secure transactions. Innovations such as HTTPS protocol, two-factor authentication, and advanced encryption techniques have bolstered the overall security of web applications, fostering trust and confidence among users. Hence, recent progress in web development has propelled the industry forward, enabling developers to craft innovative and immersive digital experiences. From responsive design to AI integration and enhanced security, the landscape of web development continues to evolve, promising a future filled with endless possibilities.

Keywords: Web development, software testing, progressive web apps, web and mobile native application.

Authors: Ian A. McCulloh, Madelaine Zinser, Jesse Patsolic, Michael Ramos

Abstract:

Recommender systems are a common artificial intelligence (AI) application. For any given input, a search system will return a rank-ordered list of similar items. As users review returned items, they must decide when to halt the search and either revise search terms or conclude their requirement is novel with no similar items in the database. We present a statistically designed experiment that investigates the impact of similarity ratings on human judgement to conclude a search item is novel and halt the search. In the study, 450 participants were recruited from Amazon Mechanical Turk to render judgement across 12 decision tasks. We find the inclusion of ratings increases the human perception that items are novel. Percent similarity increases novelty discernment when compared with star-rated similarity or the absence of a rating. Ratings reduce the time to decide and improve decision confidence. This suggests that the inclusion of similarity ratings can aid human decision-makers in knowledge search tasks.

Keywords: Ratings, rankings, crowdsourcing, empirical studies, user studies, similarity measures, human-centered computing, novelty in information retrieval.

Authors: Adam Gorine, Faten Spondon

Abstract:

Python is considered the most popular programming language and offers its own ecosystem for archiving and maintaining open-source software packages. This system is called the Python Package Index (PyPI), the repository of this programming language. Unfortunately, one-third of these software packages have vulnerabilities that allow attackers to execute code automatically when a vulnerable or malicious package is installed. This paper contributes to large-scale empirical studies investigating security issues in the Python ecosystem by evaluating package vulnerabilities. These provide a series of implications that can help the security of software ecosystems by improving the process of discovering, fixing, and managing package vulnerabilities. The vulnerable dataset is generated using the NVD, the National Vulnerability Database, and the Snyk vulnerability dataset. In addition, we evaluated 807 vulnerability reports in the NVD and 3900 publicly known security vulnerabilities in Python Package Manager (Pip) from the Snyk database from 2002 to 2022. As a result, many Python vulnerabilities appear in high severity, followed by medium severity. The most problematic areas have been improper input validation and denial of service attacks. A hybrid scanning tool that combines the three scanners, Bandit, Snyk and Dlint, which provide a clear report of the code vulnerability, is also described.

Keywords: Python vulnerabilities, Bandit, Snyk, Dlint, Python Package Index, ecosystem, static analysis, malicious attacks.

Authors: Sofia Stoica

Abstract:

In this paper, we use historical data to predict the stock price of a tech company. To this end, we use a dataset consisting of the stock prices over the past five years of 10 major tech companies: Adobe, Amazon, Apple, Facebook, Google, Microsoft, Netflix, Oracle, Salesforce, and Tesla. We implemented and tested three models – a linear regressor model, a k-nearest neighbor model (KNN), and a sequential neural network – and two algorithms – Multiplicative Weight Update and AdaBoost. We found that the sequential neural network performed the best, with a testing error of 0.18%. Interestingly, the linear model performed the second best with a testing error of 0.73%. These results show that using historical data is enough to obtain high accuracies, and a simple algorithm like linear regression has a performance similar to more sophisticated models while taking less time and resources to implement.

Keywords: Finance, machine learning, opening price, stock market.

Authors: Guezouli Larbi, Belkacem Soundes

Abstract:

The automation of the text detection process can help the human in his driving task. Its application can be very useful to help drivers to have more information about their environment by facilitating the reading of road signs such as directional signs, events, stores, etc. In this paper, a system consisting of two stages has been proposed. In the first one, we used pseudo-Zernike moments to pinpoint areas of the image that may contain text. The architecture of this part is based on three main steps, region of interest (ROI) detection, text localization, and non-text region filtering. Then, in the second step, we present a convolutional neural network architecture (On-Road Text Detection Network - ORTDN) which is considered as a classification phase. The results show that the proposed framework achieved ≈ 35 fps and an mAP of ≈ 90%, thus a low computational time with competitive accuracy.

Keywords: Text detection, CNN, PZM, deep learning.

Authors: Yuri Boiko, Iluju Kiringa, Tet Yeap

Abstract:

Current drainage attack suggested previously is further studied in regular settings of closed-loop controlled Brushless DC (BLDC) motor with Kalman filter in the feedback loop. Modeling and simulation experiments are conducted in a MATLAB environment, implementing the closed-loop control model of BLDC motor operation in position sensorless mode under Kalman filter drive. The current increase in the motor windings is caused by the controller (p-controller in our case) affected by false data injection of substitution of the angular velocity estimates with distorted values. Operation of multiplication to distortion coefficient, values of which are taken from the distortion function synchronized in its periodicity with the rotor’s position change. A saw function with a triangular tooth shape is studied herewith for the purpose of carrying out the bias injection with current drainage consequences. The specific focus here is on how the asymmetry of the tooth in the saw function affects the flow of current drainage. The purpose is two-fold: (i) to produce and collect the signature of an asymmetric saw in the attack for further pattern recognition process, and (ii) to determine conditions of improving stealthiness of such attack via regulating asymmetry in saw function used. It is found that modification of the symmetry in the saw tooth affects the periodicity of current drainage modulation. Specifically, the modulation frequency of the drained current for a fully asymmetric tooth shape coincides with the saw function modulation frequency itself. Increasing the symmetry parameter for the triangle tooth shape leads to an increase in the modulation frequency for the drained current. Moreover, such frequency reaches the switching frequency of the motor windings for fully symmetric triangular shapes, thus becoming undetectable and improving the stealthiness of the attack. Therefore, the collected signatures of the attack can serve for attack parameter identification via the pattern recognition route.

Keywords: Bias injection attack, Kalman filter, BLDC motor, control system, closed loop, P-controller, PID-controller, current drainage, saw-function, asymmetry.

Authors: Qiuyu Dai, Haochong Zhang, Xiangrong Liu

Abstract:

Efficient matrix-vector multiplication with diagonal sparse matrices is pivotal in a multitude of computational domains, ranging from scientific simulations to machine learning workloads. When encoded in the conventional Diagonal (DIA) format, these matrices often induce computational overheads due to extensive zero-padding and non-linear memory accesses, which can hamper the computational throughput, and elevate the usage of precious compute and memory resources beyond necessity. The ’DIA-Adaptive’ approach, a methodological enhancement introduced in this paper, confronts these challenges head-on by leveraging the advanced parallel instruction sets embedded within Machine Learning Units (MLUs). This research presents a thorough analysis of the DIA-Adaptive scheme’s efficacy in optimizing Sparse Matrix-Vector Multiplication (SpMV) operations. The scope of the evaluation extends to a variety of hardware architectures, examining the repercussions of distinct thread allocation strategies and cluster configurations across multiple storage formats. A dedicated computational kernel, intrinsic to the DIA-Adaptive approach, has been meticulously developed to synchronize with the nuanced performance characteristics of MLUs. Empirical results, derived from rigorous experimentation, reveal that the DIA-Adaptive methodology not only diminishes the performance bottlenecks associated with the DIA format but also exhibits pronounced enhancements in execution speed and resource utilization. The analysis delineates a marked improvement in parallelism, showcasing the DIA-Adaptive scheme’s ability to adeptly manage the interplay between storage formats, hardware capabilities, and algorithmic design. The findings suggest that this approach could set a precedent for accelerating SpMV tasks, thereby contributing significantly to the broader domain of high-performance computing and data-intensive applications.

Keywords: Adaptive method, DIA, diagonal sparse matrices, MLU, sparse matrix-vector multiplication.

Authors: Samiah Alammari, Nassim Ammour

Abstract:

When providing a massive number of tasks successively to a deep learning process, a good performance of the model requires preserving the previous tasks data to retrain the model for each upcoming classification. Otherwise, the model performs poorly due to the catastrophic forgetting phenomenon. To overcome this shortcoming, we developed a successful continual learning deep model for remote sensing hyperspectral image regions classification. The proposed neural network architecture encapsulates two trainable subnetworks. The first module adapts its weights by minimizing the discrimination error between the land-cover classes during the new task learning, and the second module tries to learn how to replicate the data of the previous tasks by discovering the latent data structure of the new task dataset. We conduct experiments on hyperspectral image (HSI) dataset on Indian Pines. The results confirm the capability of the proposed method.

Keywords: Continual learning, data reconstruction, remote sensing, hyperspectral image segmentation.

Authors: Frode van der Laak

Abstract:

The paper contributes to the feasibility of using a 5G mobile connection as a node for a Proof of Authority (PoA) blockchain, which is used for microtransactions at the same time. It uses the phone number identity of the users that are linked to the crypto wallet address. It also proposed a consensus protocol based on PoA blockchain; PoA is a permission blockchain where consensus is achieved through a set of designated authority rather than through mining, as is the case with a Proof of Work (PoW) blockchain. This report will first explain the concept of a PoA blockchain and how it works. It will then discuss the potential benefits and challenges of using a 5G mobile connection as a node in such a blockchain, and finally, the main open problem statement and proposed solutions with the requirements.

Keywords: 5G, mobile, connection, node, PoA, blockchain, microtransaction.