One of the ongoing technologies which are used by large number of companies and users is cloud computing environment. This computing technology has proved that it provides certainly a different level of efficiency, security, privacy, flexibility and availability to its users. Cloud computing delivers on demand services to the users by using various service-based models. All these models work on utility-based computing such that users pay for their used services. Along with the various advantages of the cloud computing environment, it has its own limitations and problems such as efficient resource identification or discovery, security, task scheduling, compliance and sustainability. Among these resource identification and scheduling plays an important role because users always submits their jobs and expects responses in least possible time. Research is happening all around the world to optimize the response time, make span so as to reduce the burden on the cloud resources. In this paper, QoS based service ranking model is proposed for cloud computing environment to find the essential top ranked services. Proposed model is implemented in two phases. In the first phase, similarity computation between the users and their services is considered. In the second phase, computing the missing values based on the computed similarity measures is calculated. The efficiency of the proposed ranking is measured and the average precision correlation of the proposed ranking measure is showing better results than the existing measures.
2 October 2023 in ICST Transactions on Scalable Information Systems
The most significant constraint in cloud computing infrastructure is the job/task scheduling which affords the vital role of efficiency of the entire cloud computing services and offerings. Job/ task scheduling in cloud infrastructure means that to assign best appropriate cloud resources for the given job/task by considering of different factors: execution time and cost, infrastructure scalability and reliability, platform availability and throughput, resource utilization and makespan. The proposed enhanced task scheduling algorithm using particle swarm optimization considers optimization of makespan and scheduling time. We propose the proposed model by using dynamic adjustment of parameters with discrete positioning (DAPDP) based algorithm to schedule and allocate cloud jobs/tasks that ensues optimized makespan and scheduling time. DAPDP can witness a substantial role in attaining reliability in by seeing the available, scheduled and allocated cloud resources. Our approach DAPDP compared with other existing particle swarm and optimization job/task scheduling algorithms to prove that DAPDP can save in makespan, scheduling and execution time.
27 September 2019 in Smart Innovation, Systems and Technologies
Introduction: With the rapid evolution and development of artificial intelligence and intelligent learning, the creation of realistic deepfake multimedia content has become accessible and is raising substantial requirements for digital security and media authenticity. Methods: While prevailing methods rely profoundly on deep learning and transformer driven practices, their computational cost, resource usage and sensitivity towards dataset bias prevent real-world usage and deployment. This work studies several practices for perceiving deepfake content in images and videos, analyzing state-of-the-art techniques, Convolutional Neural Network, Xception, ResNet50 and propose hybrid approach (DAAL-NET) with lightweight, Bi-stream artifact-resistant deepfake content detection capabilities to simultaneously learn spatial patterns, cues, and temporal motion inconsistencies. The framework combines three significant novelties: (1) a Local Forensics Encoder with Learnable Frequency Attention mechanism to analyze high-frequency manipulation; (2) a Motion Irregularity Encoder with depth wise temporal convolutions and gated recurrent units to obtain frame-level motion gaps; and (3) a Multi-Stream Interaction Module for bidirectional spatial temporal fusion using cross-attention. A scientifically trained Artifact Confidence Calibration Layer is proposed to improve probability and reliability. Results and discussion: Experiments supervised on Datasets of Celeb- DF(v2) and Kaggle exhibit that the proposed hybrid approach enhances macro- F1, calibration error, and temporal robustness compared to baseline models. The proposed model obtains a competitive outcome under constrained computational resources, making it appropriate for forensic applications, real-world media authentication systems, low-power deployments, and scalable deepfake screening pipelines.