In this article, we propose a novel two-phase approach to solve the Dynamic Ambulance Routing Problem (DARP), where a significant number of injured persons from different regions require treatment and medical assistance. In such cases, many people call for ambulances, but the availability and number of ambulances are insufficient to reach all patients simultaneously. Consequently, managing the ambulance fleet to respond to all demands as quickly as possible is a crucial research area to explore. In this research, we consider two main types of patients (Hard Emergency Injury [HEI] and Soft Emergency Injury [SEI]) in a dynamic context, where new demands may arise after the ambulance service has started. Furthermore, a mathematical model is proposed to formulate the DARP as a multi-objective problem that minimizes both the travel distance and ride time. To solve this NP-hard problem, we propose a two-phase approach, called k -means-MOPSO, which combines k -means clustering to group the injuries into geographic classes with the Multi-Objective Particle Swarm Optimization (MOPSO) heuristic to route the ambulances. To demonstrate the effectiveness of the k -means-MOPSO approach, this approach is compared with well-known state-of-the-art methods (NSGA-III, NSGA-II, and SPEA2) using various Pareto front metrics, such as Hypervolume, Spacing, and R2 Indicator.

Machine Learning (ML) is receiving unprecedented hype and attention. However, the ML runtime environment is still at risk from threats, such as manipulation of model parameters or contradictory poisoning of training datasets. A blockchain is a technology that combines a set of existing techniques, protocols, and tools to form a distributed and secure ledger of all transactions. This article examines and proposes a way of integrating ML suitable for Blockchain to protect the training dataset and model parameters. Another major contribution of this work is the deployment and securing of the decision process of ML, Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP) models. This smart contract-based deployment has equipped the Blockchain-based system to detect toxic recordings intelligently. The effectiveness of this proposed approach is measured in both its detection capabilities and its operational efficiency, by applying a case study of medical records as a sensitive area that tested the performance of this approach.

With the growth in services supplied over the internet, network infrastructure has become more exposed to cyber-attacks, particularly Distributed Denial of Service (DDoS) attacks, which can easily cause the disruption of services. The key factor for fighting against these attacks is the earlier separation and detection of the traffic in networks. In this paper, a novel approach, named Half Autoencoder-Stacked DNNs (HAE-SDNN) model, is proposed. We suggest using a Stacked Deep Neural Networks (SDNN) model. as a deep learning model, in order to detect DDoS attacks. Our approach allows feature selection from a preprocessed dataset using a Half AutoEncoder (HAE), resulting in a final set of important features. These features are subsequently used to train the DNNs that are stacked together by applying Softmax layer to combine their outputs. Experiments were performed on a benchmark cybersecurity dataset, named CICDDoS2017, containing various DDoS attack types. The experimental results demonstrate that the introduced model attained an overall accuracy rate of 99.95%. Moreover, the HAE-SDNN model outperformed existing models, highlighting its superiority in accurately classifying attacks.

Abstract Artificial intelligence (AI) techniques implemented at a large scale in intelligent transport systems (ITS), have considerably enhanced the vehicles' autonomous behaviour in making independent decisions about cyber threats, attacks, and faults. While, AI techniques are based on data sharing among the vehicles, it is important to note that sensitive data cannot be shared. Thus, federated learning (FL) has been implemented to protect privacy in vehicles. On the other hand, the integrity of data and the safety of aggregation are ensured by using blockchain technology. This paper applied classification approaches to VANET and ITS cyber‐threats detection at the vehicle. Subsequently, by using blockchain and by applying an aggregation strategy to different models, models from the previous step were uploaded in a smart contract. Lastly, we returned the updated models to the vehicles. Furthermore, we conducted an experimental study to measure the effectiveness of the proposed prototype. In this paper, the VeReMi data set was distributed in a balanced manner into five parts in the experimental study. Thus, classification techniques were executed by each vehicle separately, and models were generated. Upon the aggregation of the models in blockchain, they were returned to the vehicles. Lastly, the vehicles updated their decision functions and accessed the precision and accuracy of cyber‐threat detection. The results indicated that the precision and accuracy decreased by 7.1% on average with comparable F1‐score and recall. Our solution ensures the privacy preservation of vehicles whereas blockchain guarantees the safety of aggregation technique and low gas consumption.

Arabic text recognition is a difficult task due to the cursive nature of the Arabic writing system, the different forms of Arabic characters in words, the large number of ligatures, and many other challenges. Deep Learning models have made significant progress in many fields, especially in the field of Optical Characters Recognition (OCR). This article presents a model capable of recognizing handwritten Arabic characters based on deep learning. The proposed model uses Convolutional Neural Networks (CNNs) to divide the 28 Arabic characters into subclasses to improve the classification phase in OCR. The model was tested on the Handwritten Arabic Characters Database (HACDB) dataset and it gave 98% of recognition rates.

Abstract Since recording the first case in Wuhan in November 2020, COVID-19 is still spreading widely and rapidly affecting the health of millions all over the globe. For fighting against this pandemic, numerous strategies have been made, where the early isolation is considered among the most effective ones. Proposing useful methods to screen and diagnose the patient’s situation for the purpose of specifying the adequate clinical management represents a significant challenge in diminishing the rates of mortality. Inspired from this current global health situation, we introduce a new autonomous process of decision-making that consists of two modules. The first module is the data analysis based on Bayesian network that is employed to indicate the coronavirus symptoms severity and then classify COVID-19 cases as severe, moderate or mild. The second module represents the decision-making based on association rules method that generates autonomously the adequate decision. To construct the model of Bayesian network, we used an effective method-oriented data for the sake of learning its structure. As a result, the algorithm accuracy in making the correct decision is 30% and in making the adequate decision is 70%. These experimental results demonstrate the importance of the suggested methods for decision-making.

The target of a Wireless Sensor Network (WSN) designer is to improve the robustness of the network while taking into account constraints such as verifying the detection of anomalies. Preventing anomalies can be assured by data analysis via outliers identification as it strongly affects decision-making in the case of smart-home, e-health and other high-risk applications. In this paper, we study some families of outliers and we propose to detect data anomalies using K Nearest Neighbours (KNN) and Voting techniques by a benchmark suitable to be a simulation for a smart-house. In order to evaluate the dependability, some performance metrics are analyzed such as data rate detection, precision, specificity and accuracy. Performance results are in the order of 100% detection rate with reduced false alarm and fast response time compared to some other existing ones.

Social media has become an important aspect of our lives since it has infiltrated many crucial sectors including education, healthcare, entertainment and exploration. In turn, this expansion has given rise to the popularity of several online social media applications like Twitter, YouTube and Facebook in social networking. People like to share their private information with each other and they do not know what could possibly happen when they willingly disclose this kind of information or how they could prevent unwanted disclosure of their personal data. There are indeed many users who are unaware of the serious risk to the private data they publish on social media as they do not realize that these information may be used for harmful purposes and that the privacy of their data may be violated and tampered with. Since it is easier to identify social media users through their multimedia data than via any other type of data, multimedia content represents the greatest source of risk because it essentially contains pictures, videos and audio clips that jeopardize their privacy and make them more vulnerable to harm. In this paper, we will discuss the threats of the social media sites that may target the data of the users, we will also tackle the issue of privacy and security in these sites and we will discuss some of the rules and proposed solutions for protecting the data of the users of the aforementioned sites.

Vehicular ad hoc Networks (VANETs) is the most great network in size which is affecting our daily life. Safety is one of the major concern in this kind of network. Autonomous communications between vehicles help to reduce traffic jam as well as accident. In one hand, the broadcasting, as a subtype of communication, is the task of sending a message to all vehicle within a predefined distance. On the other hand, alerting other vehicles can help in reducing accidents and avoiding loses. For that reason, broadcasting techniques are used in this context and it is also used to deal with other utilities like entertainment. In this paper, we perform a comparative evaluation study on the recent proposed broadcasting techniques in VANET. Three basic categories are studied (1) based on statistical method, (2) based on Deep Learning (DL), and (3) based on mixing between cloud and DL. Finally, the future trends of broadcasting techniques in VANET are discussed.

Recently, Internet of things (IoT) is invading our daily life, which make it very attractive to industry as well as research community. the functionality of devices is prone to many failures. Discovering these failures is challenging problem due to field of devices deployment or due to device itself as resource constrained. We propose, in this paper, a novel IoT fault detection method. Feature extraction and classification is done by Support Vector Machines. Grey Wolf Optimiser (GWO) is added to the scheme for eliminating the irrelevant and redundant features. GWO optimizer is used to maximize the classification accuracy and to evaluate the selected features for the SVM classifier. This added component reflects the robustness of the proposed solution that has achieved very satisfying results. The overall accuracy is improved reaching 90.28 %.

The rise of large-scale wireless sensor networks (LSWSNs), containing thousands of sensor nodes (SNs) that spread over large geographic areas, necessitates new Quality of Service (QoS) efficient data collection techniques. Data collection and transmission in LSWSNs are considered the most challenging issues. This study presents a new hybrid protocol called MDC-K that is a combination of the K-means machine learning clustering algorithm and mobile data collector (MDC) to improve the QoS criteria of clustering protocols for LSWSNs. It is based on a new routing model using the clustering approach for LSWSNs. These protocols have the capability to adopt methods that are appropriate for clustering and routing with the best value of QoS criteria. Specifically, the proposed protocol called MDC-K uses machine learning K-means clustering algorithm to reduce energy consumption in cluster head (CH) election phase and to improve the election of CH. In addition, a mobile data collector (MDC) is used as an intermediate between the CH and the base station (BS) to further enhance the QoS criteria of WSN, to minimize time delays during data collection, and to improve the transmission phase of clustering protocol. The obtained simulation results demonstrate that MDC-K improves the energy consumption and QoS metrics compared to LEACH, LEACH-K, MDC maximum residual energy leach, and TEEN protocols.

A mobile Ad-hoc Network (MANET) become the one of top area of research, due to their simplicity of deployment. No infrastructure is required for nodes to connect with each other in the network, and with out aid of centralized administration. The nodes with MANET are quickly deployable, independent and self repair. Despite these advantages routing protocols suffering from many problems like mobility, synchronization, localization, long route and other while routing. Therefore this protocols should be study in depth, simulated in different conditions and classified. this classification and simulation helps in understanding, comparing performances and assist researchers to differentiate the characteristics and define the pros and cons of routing protocols. A detailed study and simulation model using Network Simulator (NS-2.34) with different traffic models are presented in this paper. The simulation and the performance study will focus on the impact of the network size, average energy consumption per received packet and the network density.