In the era of crowd delivery vehicle routing and traffic management in smart cities, a complex challenge appears indistinctly, affecting both developed and developing nations worldwide. This challenging problem involves optimizing multi-depot routes while addressing various hurdles: minimizing travel time, distance, fuel consumption, and carbon emissions, all while navigating dynamic traffic congestion across diverse pathways. Existing approaches often focus on isolated aspects like shortest paths, carbon emissions, or traffic prediction, leaving the comprehensive multi-depot traffic management problem unaddressed. In response, this research work proposes an Intelligent Multi-Depot Vehicle Routing and Management (IM-VRM) model which provides a comprehensive and holistic solution. It employs a Graph Neural Network (GNN) learning-based routing with a greedy optimization to establish initial optimal pathways for multi-depot journeys. Subsequently, the IM-VRM model integrates traffic congestion prediction with green parameter computation, engaging the Dijkstra algorithm to select the most admissible routes. This consecutive steps-based travel route guidance process optimizes routing for heterogeneous vehicles, including both heavy-duty and light-duty types. It accounts for load-dependent fuel consumption, velocity, and carbon emissions. By doing so, it simplifies the complexities of multi-depot traffic routing and management. The proposed model has been rigorously evaluated using a real-world multi-depot traffic dataset, demonstrating its practical viability. Notably, IM-VRM model achieves a remarkable improvement in fuel savings, reduced carbon emissions, and shorter travel time outperforming previous state-of-the-art methods in both efficiency and precision.

Pedestrian trajectory prediction aims to predict future trajectories based on observed trajectories. Current state-of-the-art methods often assume that the observed sequences of agents are complete, which is a strong assumption that overlooks inherent uncertainties. Understanding pedestrian behavior when dealing with missing values in the observed sequence is crucial for enhancing the performance of predictive models. In this work, we propose the MultiScale hypergraph for Trajectory Imputation and Prediction (MS-TIP), a novel approach that simultaneously addresses the imputation of missing observations and the prediction of future trajectories. Specifically, we leverage transformers with diagonal masked self-attention to impute incomplete observations. Further, our approach promotes complex interaction modeling through multi-scale hypergraphs, optimizing our trajectory prediction module to capture different types of interactions. With the inclusion of scenic attention, we learn contextual scene information, instead of sole reliance on coordinates. Additionally, our approach utilizes an intermediate control point and refinement module to infer future trajectories accurately. Extensive experiments validate the efficacy of MS-TIP in precisely predicting pedestrian future trajectories. Code is publicly available at https://github.com/Pranav-chib/MS-TIP.

The key challenges in cloud computing encompass dynamic resource scaling, load balancing, and power consumption. Accurate workload prediction is identified as a crucial strategy to address these challenges. Despite numerous methods proposed to tackle this issue, existing approaches fall short of capturing the high-variance nature of volatile and dynamic cloud workloads. Consequently, this paper introduces a novel model aimed at addressing this limitation. This paper presents a novel Multiple Controlled Toffoli-driven Adaptive Quantum Neural Network (MCT-AQNN) model to establish an empirical solution to complex, elastic as well as challenging workload prediction problems by optimizing the exploration, adaption, and exploitation proficiencies through quantum learning. The computational adaptability of quantum computing is ingrained with machine learning algorithms to derive more precise correlations from dynamic and complex workloads. The furnished input data point and hatched neural weights are refitted in the form of qubits while the controlling effects of Multiple Controlled Toffoli (MCT) gates are operated at the hidden and output layers of Quantum Neural Network (QNN) for enhancing learning capabilities. Complimentarily, a Uniformly Adaptive Quantum Machine Learning (UAQL) algorithm has evolved to functionally and effectually train the QNN. The extensive experiments are conducted and the comparisons are performed with state-of-the-art methods using four real-world benchmark datasets. Experimental results evince that MCT-AQNN has up to 32%-96% higher accuracy than the existing approaches.

Cloud virtualization technology, ingrained with physical resource sharing, prompts cybersecurity threats on users’ virtual machines (VMs) due to the presence of inevitable vulnerabilities on the offsite servers. Contrary to the existing works which concentrated on reducing resource sharing and encryption/decryption of data before transfer for improving cybersecurity which raises computational cost overhead, the proposed model operates diversely for efficiently serving the same purpose. This article proposes a novel multiple risks analysis-based VM threat prediction model (MR-TPM) to secure computational data and minimize adversary breaches by proactively estimating the VMs threats. It considers multiple cybersecurity risk factors associated with the configuration and management of VMs, along with analysis of users’ behavior. All these threat factors are quantified for the generation of respective risk score values and fed as input into a machine learning-based classifier to estimate the probability of threat for each VM. The performance of MR-TPM is evaluated using benchmark Google Cluster and OpenNebula VM threat traces. The experimental results demonstrate that the proposed model efficiently computes the cybersecurity risks and learns the VM threat patterns from historical and live data samples. The deployment of MR-TPM with existing VM allocation policies reduces cybersecurity threats up to 88.9%.

Cloud computing has become inevitable for every digital service which has exponentially increased its usage. However, a tremendous surge in cloud resource demand stave off service availability resulting into outages, performance degradation, load imbalance, and excessive power-consumption. The existing approaches mainly attempt to address the problem by using multi-cloud and running multiple replicas of a virtual machine (VM) which accounts for high operational-cost. This paper proposes a Fault Tolerant Elastic Resource Management (FT-ERM) framework that addresses aforementioned problem from a different perspective by inducing high-availability in servers and VMs. Specifically,(1)an online failure predictor is developed to anticipate failure-prone VMs based on predicted resource contention;(2)the operational status of server is monitored with the help of power analyser, resource estimator and thermal analyser to identify any failure due to overloading and overheating of servers proactively; and(3)failure-prone VMs are assigned to proposed fault-tolerance unit composed of decision matrix and safe box to trigger VM migration and handle any outage beforehand while maintaining desired level of availability for cloud users. The proposed framework is evaluated and compared against state-of-the-arts by executing experiments using two real-world datasets. FT-ERM improved the availability of the services up to 34.47% and scales down VM-migration and power-consumption up to 88.6% and 62.4%, respectively over without FT-ERM approach.

Data leakage has become a critical challenge for any organisation, especially when the third party is involved with legitimate permissions to access its confidential data. This study provides a statistical evaluation‐based leaker identification (SELI) scheme to identify the malicious entity when the data is leaked by any agent at some unauthorised place. A new distribution strategy for object and agent selection which enriches the guilty agent recognition is introduced. The three algorithms named SELI–first come first serve, SELI–round Robin (SELI–RR) and SELI–shortest request first (SELI–SRF) are presented for the agent selection. Furthermore, the spurious data objects are added while distributing the data sets among the agents to uniquely identify the leaker. The SELI scheme estimates the guilty entity based on the allocated data among the agents using bigraph. The proposed solution achieves an improvement of up to 25.37 and 58.18% for average probability and average success rate respectively in case of SELI–SRF and 163.68% in case of SELI–RR for detection rate compared to the prior work. In addition to this, the SELI scheme proved its efficiency by securing up to 99.82% accuracy, 99.92% precision, 99.4% recall and 99.97% specificity.

Data sharing across multiple different entities is on‐demand to upgrade an enterprise's performance. However, some malicious entity can reveal this data to an unauthorised third party that may result in heavy loss to the enterprises in terms of finance, reputation, and long‐term stability. This study presents a novel model GUIM‐SMD for the identification of the guilty entity which is responsible for the data leakage to the unauthorised party in the shared environment. An effective distribution strategy to allocate the data among the users based on the access control mechanism is proposed in this model. The approach introduces the summation matrix which is computed using D‐score and U‐score that are assigned to the classified data and user, correspondingly. Furthermore, D‐score and U‐score are based on the data sensitivity and user guilty record relatively; and their values vary between 0 and 1. The evaluated summation matrix is used for data distribution among various users. The results show improvement up to 98.74, 236.38, and 252.39% for average probability, average success rate, and detection rate, respectively, as compared to the prior work.