Cloud Computing is the Technology that changed the entire world by providing services on-demand through the Internet. Various service providers give service at different levels of Cloud Models to customers on a pay-per-usage basis. Customers get those services independent of their locations. One of the major challenges of Cloud technology is managing and provisioning the available resources dynamically. Various measures were proposed to manage the resources efficiently by various researchers. Machine Learning is the latest trend in the area of Computer Science which focuses on data and algorithms using Artificial Intelligence concepts. In this paper, we propose two of the Machine Learning Classification approaches, Bayes Classification and Logical Regression, to achieve Dynamic Resource Management of Cloud resources. These two approaches were analyzed to predict the availability of required resources from the Resource Pool to provision the Resources very efficiently by the Cloud Service Provider in real time, according to the Service-Level agreements made with the Consumer.