Using Additive and Deep Learning Algorithms for Weather Forecasting

Abstract

One significant element that directly impacts agricultural activities is the weather. The type of crop that should be grown depends significantly on the temperature and humidity of the area. Therefore, farmers need to be aware of upcoming weather trends to plan their farming operations. Although there are several traditional weather prediction services, they are always run by governments and rely on intricate modelling. In this scenario, time series forecasting becomes an effective method of predicting these temperature patterns because it needs historical weather data and very little computer power to provide results. In this work, we examine the steps involved in putting both additive and regression-based models into practice and evaluate their respective performances to determine which strategy is most effective for weather prediction.