Cloud computing is very important for many companies in the process of progress. The main problem for any company when transferring their work to the cloud is selecting the most suitable cloud provider among the availability of different cloud service providers with different properties and different alternatives. This paper introduces a novel framework that can be used for selecting the most suitable provider in the case of missing values in the evaluation of alternatives. The framework is composed of two steps; the first step in the framework is about using the Modified Generative Adversarial Network (M-GAN) for data imputation of missing data. The modified version of GAN has achieved an accuracy of nearly 0.94. The second step is the Multi-Criteria Decision-Making (MCDM) neutrosophic algorithm for selecting the most suitable provider according to different eight criteria (Availability, Throughput, Successibility, Reliability, Latency, Response time, Response Time of Customer Services, and Cost). According to the experiments done in the paper, the Novel framework has achieved success in choosing suitable providers. the presented model achieved 0.05 (sec) computation time for 1000 providers rather than 0.057 (sec), 0.061 (sec), and 0.065 (sec) in other mentioned works.
Digital Object Identifier (DOI)
Attya, Mohammed; S. Sakr, Ahmed; M. Abdulkader, H.; Al-Showaikh, Faisal; and Kamel El-Sayed, M.
"Novel Framework for Selecting Cloud Provider Using Neutrosophic and Modified GAN,"
Applied Mathematics & Information Sciences: Vol. 17:
2, Article 12.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol17/iss2/12