학술논문
Cloud Risk Management With OWA-LSTM and Fuzzy Linguistic Decision Making
Document Type
Periodical
Author
Source
IEEE Transactions on Fuzzy Systems IEEE Trans. Fuzzy Syst. Fuzzy Systems, IEEE Transactions on. 30(11):4657-4666 Nov, 2022
Subject
Language
ISSN
1063-6706
1941-0034
1941-0034
Abstract
In a cloud environment, the indemnity of service level agreement (SLA) violations has an adverse effect on the service provider. It leads to the penalty fee, credit amount, license extension, and reputation decline that could significantly impact future business outcomes. Existing approaches are unable to handle complex predictions that can accommodate the temporal influence of Quality of Service (QoS) data. Moreover, no method in a cloud environment considers all possible attitudinal behavior of the service provider to mitigate the risk of an actual violation. This article proposes an SLA violation risk mitigation model that uses ordered weighted average (OWA) in long short-term memory for complex QoS prediction. The OWA operator is weighted with a minimax disparity approach to manage the risk of SLA violation. The approach intelligently predicts deviation in custom prioritized QoS parameter and recommend exigency of mitigating action by considering all possible attitudinal behavior of the service provider. This article uses linguistic variables, fuzzy and interval numbers to handle imprecise information. The analysis results demonstrate the applicability and efficiency of the proposed approach to address complex risk mitigation actions.