A Markov Chain-Based Multi-Criteria Framework for Dynamic Cloud Service Selection Using User Feedback

Abstract

As the variety of cloud services continues to expand, organizations increasingly struggle to select the best options. This task is made more challenging by the ever-changing nature of user preferences, which shift based on evolving needs and feedback from previous service experiences. This paper tackles these issues by presenting a comprehensive multi-criteria decision-making (MCDM) framework to aid in cloud service selection. The framework utilizes a Markov chain model to analyze and discern patterns in user feedback, facilitating the ranking of cloud services based on both quality and user satisfaction metrics. By employing a Markov chain approach, the framework can track changes in user preferences over time, providing a dynamic means of evaluating cloud services. This system aids users in making informed choices by offering personalized recommendations that meet their specific needs and preferences. It also provides cloud service providers (CSPs) with valuable insights into market trends and customer expectations, helping them enhance their services. The framework's efficacy is demonstrated through a detailed simulation using real-world quality of service (QoS) data. Furthermore, a comprehensive sensitivity analysis is performed to assess the robustness and reliability of the proposed approach, ensuring its consistency.