The Covid-19 pandemic has left the world in great adversity. During this situation, every region has various reasons for the spread of the COVID-19 virus. Although various countries have executed different strategies- pharmaceutical and non-pharmaceutical- they are unable to control the spread of the virus. This shows one cannot determine the correct method for controlling the spread and flattening the curve of the COVID-19 cases. However, using machine learning, we can surely rule out the strategies/methods which are not effective. This can be efficiently carried out by collecting time-series data of a particular country/region and analyse it using appropriate Neural Network architecture to predict the curve. In this paper, we demonstrate a method that satisfactorily predicts the virus spread in the Maharashtra state of India. The paper also shows a method to predict the virus spread in any region based on various parameters and decisions taken by governing authority along with the impact of that decision.