With the deployments of fifth generation mobile networks (5G), rapid development of mobile internet, continued growth in mobile traffic and increased adoption of the internet of things, Multi-access Edge Computing (MEC) remains an inevitable critical future radio access network artefact for an optimized network. Network operators as part of their optimization exercise are also adopting Cloud Radio Access Network (C-RAN) to further improve network operation performance. Combining MEC and C-RAN makes the economics (OPEX + CAPEX) of both technologies more attractive and also enables network operators to support key 5G applications. This research investigates a major challenge of combining MEC + C-RAN: how to efficiently orchestrate cloud resources as a user moves within the edge environment without compromising the overall quality of service. This ongoing research adopts a machine learning technique in orchestrating resources efficiently in order for applications and services to adhere to stringent performance requirements even at geographically dispersed "edge" locations.