International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE)
Not a member yet
85 research outputs found
Sort by
A Scheduling Algorithm based on PSO Heuristic in Cloud Computing
No full textThe goal of the SaaS provider is the most protable; the user\u27s goal is to meet requirements as quickly as possible but still within budget and deadline. The algorithm\u27s aim gives the schedule to satisfy the objectives of the agents, this is a very difficult problem. This article studies heuristic PSO (Particle Swarm Optimization) and model of components in the cloud computing to propose a model of PaaS providers; admission control algorithm and scheduling for the user’s requirements towards multi-objective optimization of time.The schedule given by the algorithm in order to: (1) optimizing the time for the user, (2) providing the greatest benefits for SaaS providers, (3) satisfying for the constraints of QoS (Quality of Service ) of the user. The result of the algorithm is installed and compared with other algorithms on CloudSim
A Review on Cloud Computing Security and Privacy in Service Oriented Architecture (SOA)
No full textCloud computing is powerful computing which is transformed the traditional computing and industries. Cloud is not a new concept. Many organizations like Google, Microsoft, and Amazon accelerate in developing this computing and provide the services for lots of users and storing the data through cloud now become a norm. But there are many issues that arrive to store the data in cloud. In this paper we review some securities issues and give a survey solution that have been done to minimize the security risk and describe future research work about all these risk that occur when data is stored in cloud computing
Improvement of Automated Learning Methods based on Linear Learning Algorithms
No full textIn recent years, the process of learning creatures is converted to one of the new research area. These researches are divided into two general categories that one of them is based on proposing a solution and learning based methodology to any machines. Learning is defined as changes made in the performance of a system based on experiences. The most prominent features of learning-based systems are that they improve themselves over time. Therefore, learning based machines have a big role in these systems. However, they are not very productive in some application and research areas such as smart real time systems especially. In this paper is proposed a new approach based on reinforcement learning technique that has three versions in order to implementation in different areas. It behaviors based on reward and penalty model. The effectiveness of these interactions with the environment is evaluated by the maximum (minimum) of the number of rewards (penalty) taken from the environment. The main advantage of the reinforcement learning over other learning methods is the need for no information from the environment (except amplification signal). The other learning methods as supervised or unsupervised are not appropriate to these problems. In this method, each agent decides the next its actions based on current k-actions instead of one action. The three versions are simple, sequential and unstructured linear learning methods so they evaluated in different possibilities to get the appropriate responses. Depending on the needs of any system, they can be used. The mode of convergence of actions in the proposed automaton (machine) in six different scenarios is examined.  
IoT with Big Data Framework using Machine Learning Approach
No full textIn future IoT (Internet of Things), big-data administration & machine learning disclosure for expansive scale modern robotization application, the significance of mechanical internet is expanding step by step. The interconnection by means of the Internet of computing gadgets installed in ordinary items, empowering them to send and get information. BD is informational collections that are so voluminous and complex that customary information preparing application programming are insufficient to manage them. ML is a subset of artificial intelligence that regularly utilizes measurable procedures to enable PCs to "learn" with information, without being expressly modified. A few differentiated advancements, for example, IoT, computational intelligence, machine type communication, BD, & sensor technology can be fused together to enhance the data administration & information revelation effectiveness of expansive scale robotization applications. An expanding measure of significant data sources, propels in IoT & Big Data (BD) advances & also the accessibility of an extensive variety of machine learning (ML) calculations offers new potential to convey logical administrations to nationals & urban chiefs. In any case, there is as yet a hole in joining the present best in class in an incorporated system that would help lessening improvement costs & empower new sort of administrations. Voluminous measures of data have been created, since the previous decade as the scaling down of IoT gadgets increments. Be that as it may, such data are not valuable without scientific power. Various BD, IoT, & investigation arrangements have empowered individuals to acquire profitable knowledge into extensive information created by IoT gadgets. However, these arrangements are still in their earliest stages, & the domain does not have a thorough review on this. Here we endeavored to give a reasonable more profound understanding about the IoT in BD structure alongside its different issues & challenges & concentrated on giving conceivable arrangement by ML strategy
Internet of Things and Healthcare Technologies: A Valuable Synergy from Design to Implementation
No full textInternet of Things (IoT) promises to be a reliable technology for the future. Healthcare is one of the fields which are rapidly developing new solutions. The synergy between IoT and healthcare promises to be very beneficial for human healthcare and evolved into a new field of research and development: the Internet of Medical Things (IoMT). This paper presents a review on various enabling IoMT technologies based on the latest publications and technology available in the marketplace. This article also analyzes the various software platforms available in the field of IoMT and the current challenges faced by the industr
DNA Based Storage: Introduction, Characteristics, Applications and Challenges
No full textOver the years, as humans have made progress, data has come to the forefront and has become one of the principal elements of life. No matter the field, all aspects of life are now dependent on data in one way or the other. Be it hospitals or financial institutions; sports teams or researchers, all operate on some form of data during their functioning. This ever-increasing dependency on data further leads to the need for its storage. The capability of present storage mechanisms is not able to keep up with the exponentially increasing demand. This along with other factors such as high setup costs, high maintenance charges, security, and accessibility are pushing towards an alternative avenue of storage. DNA or the code of life is very similar to the binary based data systems that we operate on, hence is being looked at, as the alternative to conventional methods. This field has seen massive amounts of developments in the recent past and is finding a strong footing. Its theoretical capability to store all the data ever created in a finger-sized device is one of the many factors, which makes it such an interesting field to study and know about. This paper describes how this domain of storage system(s) basically functions, the work is done in this field in the past, its advantages and limitations along with the challenges that this domain needs to overcome to become practically viable bringing a paradigm shift in computing
An Adaptive Load Balancing Queue Based Resource Allocation Algorithm in Cloud Computing Environment
No full textLoad balancing provides a better approach to distribute workloads at different virtual machines in different data center. In cloud computing improves the distributions of workloads across multiple cloud computing resources such as a computer, computer clusters, network links, central processing unit or disk drives. The aim of load balancing is to optimize maximum throughput, minimum response time, resource use and avoid overload on any single resource. Using multiple components with load balancing instead of single component may increase reliability and availability through redundancy. Cloud computing is one of the greatest platforms which provides high user satisfaction ratio and optimize and distribute load at different virtual machine according to client requirements
Steganalysis for Reversible Data Hiding based on Neural Networks and Convolutional Neural Networks
No full textLossless data hiding techniques is a technique that is very interested. In which there is a large amount of reversible information hidden technologies. This technique is technically possible to restore the original image after extracting the information from the stego image. The stego image (image to be hidden secret data) is not detected hardly any variable. There are many studies for this field is published. Secret information is hidden on the pixel space, frequency (cosine, wavelet) coefficient space or difference image coefficient space. However, by analysing meticulously between the cover image and the stego image on these space can be detect abnormal signs. In my previous work, we produced a steganalytic techniques based on analysing the transform coefficient histogram with the correct detection ratio between 88% and 92%. In this article, my team give another method to improve the detection ratio of that steganalysis based on Neural Networks (NNs) and Convolutional Neural Networks (CNNs). Our test results show 94% correct detection rates for NNs and 93% for CNNs, this is a better result than our previous method. This proposed approach can be applied to detect stego images on spatial and other frequency domain
Dynamic Trust: A Protected and Trustable Directing in Remote Sensor Systems
No full textExpansive scale sensor structures are sent in various application ranges, and the information they gather are utilized as a bit of major specialist for fundamental foundations. Information are gushed from different sources through transitional dealing with focus focuses that total data. A poisonous enemy may demonstrate extra focus focuses in the structure or arrangement existing ones. As necessities be, guaranteeing high information steadfast quality is basic for audit basic activity. Information provenance tends to a key factor in looking over the steady nature of sensor information. Provenance association for sensor systems demonstrates two or three testing fundamentals, for example, low centrality and data trade constrain utilization, convincing point of confinement and secure transmission. In this paper, we propose a novel lightweight game plan to safely transmit provenance for sensor information. The proposed procedure depends upon in-dispense channels to encode provenance. We present convincing instruments for provenance certification and expansion at the base station. Plus, we expand the secured provenance plot with handiness to see isolate assaults coordinated by pernicious information sending focus focuses. We assess the proposed structure both effectively and absolutely, and the outcomes demonstrate the attainability and ability of the lightweight secure provenance plot in seeing pack produce and calamity assaults
Multidimensional Performance analysis for Packet delivery and routing overhead in AODV and AOMDV
No full textA Mobile Ad-hoc Network is characterized as a system that is remote and dynamic. It can be outlined with no need for earlier framework where each hub goes about as switch. A versatile Ad hoc Network is a self-arranging arrangement of portable hubs that are associated remotely. Each hub capacities as a sink and in addition a switch to send packets. These hubs can move unreservedly and freely toward any path and ready to get sorted out into a system. Thus, they change their positions as often as possible. In this study, a correlation is made between Ad-hoc On Demand Distance Vector convention and Ad-hoc On Multipath Demand Distance Vector convention utilizing system test system NS2.35.
AODV is reactive gateway discovery algorithm where a MANET mobile device connects only on-demand. AOMDV was basically made for highly dynamic ad-hoc networks to respond to link breakages and failures in network. It deals with managing ways for the goals and utilizations goal arrangement numbers to define the fresh routing to guarantee circle flexibility consistently and to stay away from issues. It is a protocol based on timer that finds ways for the mobile nodes to respond to breakages in links and changes in topology. The result demonstrates that the AODV is superior to AOMDV when the number of node increases. Then again, the AOMDV has better performance when the simulation increases