Improving Reliability of Mobile Social Cloud Computing using Machine Learning in Content Addressable Network

Abstract

Mobile social cloud computing (MSCC) is a paradigm that focuses on sharing data and services between end-users over a scalable network of cloud servers, mobile, computers, and web services. Quality of Service (QoS) based task provisioning in MSCC is one of the most eminent optimization problems, also used in improving the performance of system and efficient service delivery. Cloud based social networking service (SNS) is an application platform where individuals with like interests, family, and friends communicate with each other and share the data with less or no authentication. In MSCC, the user mobility is supported by infrastructure like access points (APs) and networking protocols. Content Addressable Network (CAN) is used to provide logical structure to resources (mobile devices and servers) and look up any resource on cloud servers. MSCC performance essentially includes QoS requirement that evaluates the quality of MSCC. Apart from basic QoS like time and cost, extended QoS is crucial for evaluating these networks. In this work, a machine learning-based framework is proposed for improving QoS of MSCC through reliability. This framework not only optimizes QoS but also restrains the malicious nodes by taking feedback from ML method.