Artificial Intelligence Based Prediction Models and Their Applications in Agriculture

Abstract

Forecasting is used to provide an aid to decision-making and in planning the future effectively and efficiently. It is an important aspect for a developing economy so that adequate planning is undertaken for sustainable development. The Box-Jenkins Autoregressive Integrated Moving Average (ARIMA) model is the most widely used classical time series model for forecasting in agricultural and allied sector. The major drawback of classical model is presumption of linearity, hence, no nonlinear patterns can be recognized by these models. Alternatively, intelligence techniques (AI) models are very effective to capture the nonlinear complex and heterogeneous pattern present in the agricultural data. Sometimes, the data under consideration contains both linear and nonlinear patterns, under such condition classical and AI models fails to capture the trend properly.  To overcome this problem, a hybrid model or two stage methodology can be employed. Some of the popular classical, AI and hybrid methodologies are described in this article.