Resource Management in the network function virtualization (NFV) environment is a challenging task. The continuously varying demands of virtual network functions (VNF) call for dynamic algorithms to efficiently scale the allocated resources and meet fluctuating needs. In this context, studying the behavior of a VNF as a function of its environment helps to model its resource requirements and thus allocate them dynamically. This paper investigates the use of machine learning (ML) techniques to estimate VNFs needs in term of CPU as a function of the traffic they will process. We propose and adapt a Support Vector Regression (SVR) based approach to resolve the problem. Results show its efficiency and superiority compared to the state of the art