Weather is a natural phenomenon which affects almost every activity of the native people and economic growth. Its dependency on the location is a known fact, however some other latent factors also affect the sudden change in weather parameters for any location. It is a difficult job for researchers to predict the weather from an environmental standpoint, and in recent years, it has garnered a lot of interest and attention. This study investigates a variety of approaches for measuring and forecasting weather that take into account the daily, monthly, or annual meteorological data. However, because of the environment's complexity and dynamic nature, it is still challenging to predict meteorological characteristics with accuracy. The prediction of air parameters is being attempted using a variety of machine learning methods. This study examines a number of applications that rely on the results of numerical weather prediction models in addition to weather forecasting.