The proliferation of malicious online behaviors, particularly cyber-trolling, presents significant challenges to maintaining healthy online communities. This paper investigates the efficacy of four deep learning architectures-BERT, LSTM, GRU, and Causal Convolutional Networks (Causal Conv 1D)-for the automatic detection of cyber-trolls based on textual content. Using a comprehensive dataset of 50,000 social media comments, we evaluate these models on their ability to distinguish between normal users and trolls. Our results indicate that while the pre-trained BERT model achieves the highest overall accuracy ($94.2 \%$), the Causal Conv 1D architecture demonstrates competitive performance $(92.7 \%)$ with significantly lower computational requirements. We also analyze the semantic features that most effectively contribute to troll detection and discuss the ethical implications of automated moderation systems. This research contributes to the development of more efficient and effective methods for maintaining civil discourse in online spaces.

This paper presents a taxonomy of existing ”secure optimization” VM placement strategies. Notably, the Approximate Co-Location-Resistant Placement Strategy with Risk Constraint (ACLRPS-RC) demonstrated distinguishable performance in terms of security, resource optimization, and risk mitigation. However, its performance was solely examined using the PlanetLab workloads. To assess the robustness of ACLRPS-RC, we evaluate the effects of varying security and risk parameters using additional datasets with diverse characteristics. The experiments provide valuable insights into the algorithm’s generalizability across different workloads.

In cloud computing, muti-tenancy enables virtual machines to co-host the same server to reduce the power expenses. Although, there may exist co-hosts that attempt to conduct co-location attacks, which cause data leakage. To address this challenge, the most recent works present Virtual machines placement strategies minimizing co-location attacks and energy consumption. Few works establish a security before-hand threshold to constraint attack probability with respect to the provider/users priorities. Thus, we adopt this approach and introduce a novel virtual machines allocation strategy that considerably reduces energy consumption. Finally, we deploy our strategy and prove its high performance on the simulation platform CloudSim.

Abstract Named data networking (NDN) has emerged as a promising communication paradigm, proposed to deal with the shortcomings of the traditional IP‐based model. NDN introduces new name‐based routing, receiver‐based service, caching, and self‐certifying contents features that obviously improve data delivery efficiency and reliability. Moreover, NDN offers lightweight forwarding rules that suits constrained devices. These features makes NDN as highly promising communication model for the Internet of Things (IoT). On the other hand, one of the widely adopted networking specification for IoT is the IEEE 802.15.4 standard. This latter proposes interesting energy saving functionalities. To take full advantages of the two technologies, we propose an NDN over IEEE 802.15.4 communication solution that meets the requirements of low‐data rate and low‐power‐consumption monitor and control IoT applications. The proposed solution includes two modules: a reliable energy‐aware forwarding strategy that selects the next hop forwarder based on its residual energy level and a sleep mode scheduling algorithm that schedules the sleep/wake‐up mode according to the role of the node in the forwarding and the path repair processes. Extensive simulations and analyses have been conducted to confirm the viability and effectiveness of our proposal in terms of energy consumption, network lifetime, delivery ratio, retrieval delay, and scalability.

VANETs are characterized by the rapid changes in network topology due to their random movement patterns and their high-speed mobility. Hence, the support of efficient mobility management solutions become an important feature in VANET. Most of VANET applications need internet access almost everywhere and at any time without interruption. Thus, ensuring a seamless connection and enhanced throughput performance requires an improved handover strategy. In this paper, we introduce a new hand over optimization method for the 5G cellular network. A mobility prediction algorithm coupled with previous handover events logs was used to predict when and where the handover will occur in the network. In the proposed work, we aim to minimize the number of handover events without degrading network performance. A simulation-based performance study was conducted to evaluate the effectiveness of the proposed methods, and the results were compared to the 3GPP conventional handover solution. It was found that our proposed solution reduces the number of handover events without affecting the network quality.

The vehicular ad-hoc networks (VANET) has attracted the attention of both the industry and the academia researcher over the last decade. The concept of connecting vehicles to the Internet using the already deployed cellular network architecture has opened many avenues for research and development that are contributing significantly towards the Intelligent Transportation Systems (ITS). Almost every vehicle requires a seamless connectivity to the Internet without interruption. However, with the emergence of the 5G network and the Vehicle-to-infrastructure(V2I) concept, the design of efficient mobility management techniques that can handle the real-world mobility constraints in VANET becomes a critical task. In this paper, we propose a new handover algorithm that uses a Markov chain predictor to determine when and where a handover will be needed. The aim of the proposed solution is to reduce the number of unnecessary handover by maintaining the vehicle connectivity to the 5G base station as long as possible without degrading the network performance. Simulation studies were conducted to evaluate the performance of the proposed scheme. Our results show that the proposed handover algorithm greatly outperforms the conventional 3GPP handover algorithms.

Falls are prevalent in the elderly population and there is an urgent need for public health strategies to decrease their incidence and identify those who are at risk. Physicians are increasingly confronted by critical and serious situations in patients at risk of falls who require effective interventions. Being aware of the patient’s medical record, they need to quickly and efficiently recommend the best therapeutic intervention to reduce the incidence of future falls. In this respect, we propose an Interventions Recommendation System that aids practitioners make decisions about elderly falls by recommending individualized intervention that would reduce the risk of patient’s future falls. This paper describes our work in progress on a probabilistic causal model for preventing falls in older adults. We conduct an initial empirical study for such a model on an elderly personal information base and report the initial promising results of our causal model in terms of the usefulness and effectiveness of our approach.

The information describing a person is constantly evolving and may become obsolete and contradict other information. A personal database, therefore, must be consistently updated upon the acquisition of new valid observations that contradict obsolete ones contained in the database. This study focuses on proposing a novel approach for dealing with the information obsolescence problem in the elderly-fall prevention context. Our approach aims to continuously monitor elderly information in order to detect the change in the behavior of an elderly person and to prevent him from falls. It consists of detecting contradictions between newly acquired information about a single elderly and what we already know about that person, then identifying among his observations those that have become obsolete and need to be updated. We propose a new approximate concept,£-Contradiction, which represents the confidence level of having a contradiction in a set of observations when a causal Bayesian network is our representation model. We propose a polynomial-time algorithm for detecting obsolete information and show that the resulting obsolete observations are given as an explanation AND-OR tree.

Power consumption and network lifetime optimization is an essential performance objective for the IoT networks. Since the devices in this type of networks are mostly battery powered with a limited battery life, energy saving strategies becomes vital for such systems operation. Typically, the standard sleep mode of the 802.15.4 protocol is used in order to reduce the power usage of the sensors in an IoT network. However, enabling the sleep mode function in an IoT device that utilizes the Named Data Networking (NDN) paradigm is challenging. In this paper, we propose a power saving extension for the Geographic Interest Forwarding (GIF) protocol. In the proposed E-GIF scheme, a cross-layer approach is used, which allows direct communication between protocols at nonadjacent layers. More precisely, the interactions between the MAC and the Routing layers are fully exploited in order to integrate the sleep mode into the forwarding decision of the sensor nodes that uses the NDN protocol stack. To validate the proposed scheme, we extend the ndnSIM simulator to support wireless multihop communication. Simulation experiments confirm the viability and the effectiveness of the proposal.

In this paper we present a new Smart Online Vehicle Tracking System for Security Applications (AMOTSSA) and we describe how it can be modelled and implemented as a Big data application. In order to model AMOTSSA as a big data application, we argue our design choices that meets its specific data and processing needs and we present a set of data analytic algorithms that would achieve a set of investigation support goals.

In constraint solving, a critical bottleneck is the formulation of aneffective constraint model of an input problem. The Conjure system describedin this paper, a substantial step forward over prototype versions of Conjurepreviously reported, makes a valuable contribution to the automation ofconstraint modelling by automatically producing constraint models from theirspecifications in the abstract constraint specification language Essence. Aset of rules is used to refine an abstract specification into a concreteconstraint model. We demonstrate that this set of rules is readily extensibleto increase the space of possible constraint models Conjure can produce. Ourempirical results confirm that Conjure can reproduce successfully the kernelsof the constraint models of 32 benchmark problems found in the literature.

We study the CARDPATH constraint. This ensures a given constraint holds a number of times down a sequence of variables. We show that SLIDE, a special case of CARDPATH where the slid constraint must hold always, can be used to encode a wide range of sliding sequence constraints including CARDPATH itself. We consider how to propagate SLIDE and provide a complete propagator for CARDPATH. Since propagation is NP-hard in general, we identify special cases where propagation takes polynomial time. Our experiments demonstrate that using SLIDE to encode global constraints can be as efficient and effective as specialised propagators.

When writing a constraint program, we have to choose which variables should be the decision variables, and how to represent the constraints on these variables. In many cases, there is considerable choice for the decision variables. Consider, for example, permutation problems in which we have as many values as variables, and each variable takes an unique value. In such problems, we can choose between a primal and a dual viewpoint. In the dual viewpoint, each dual variable represents one of the primal values, whilst each dual value represents one of the primal variables. Alternatively, by means of channelling constraints to link the primal and dual variables, we can have a combined model with both sets of variables. In this paper, we perform an extensive theoretical and empirical study of such primal, dual and combined models for two classes of problems: permutation problems and injection problems. Our results show that it often be advantageous to use multiple viewpoints, and to have constraints which channel between them to maintain consistency. They also illustrate a general methodology for comparing different constraint models.

By focusing on the families of assignment and permutation problems (such as graph colouring and n-Queens), we show how to adapt D.R. Smith's (1990) KIDS approach for the synthesis of constraint programs (with implicit constraint satisfaction code), rather than applicative Refine programs with explicit constraint propagation and pruning code. Synthesis is guided by a global search schema and can be fully automated with little effort, due to some innovative ideas. CLP (Sets) programs are equivalent in expressiveness to our input specifications. The synthesised CLP (FD) programs would be, after optimising transformations, competitive with carefully hand-crafted ones.