Predictive Analytics Techniques in the Agricultural Industry: A Case Study

Agriculture is the main occupation and backbone of the people across the globe. In recent days, many young learned people get attracted to the agriculture field. Farmers used to decide what crops to grow in the land depending on their previous experience. Every farmer strives to increase his or her crop production. Agricultural yields, on the other hand, are largely determined by weather conditions, therefore farmers may not receive the predicted harvest owing to extreme weather circumstances such as excessive rain, low rainfall, and other weather conditions. Crop yield forecast aids farmers in deciding what to produce, when to grow it, and how much to grow. Recent advances in information technology for agriculture have sparked an interest in crop yield prediction research. Crop yield prediction is a technique for predicting crop yields based on a variety of factors such as rainfall and temperature, cloud coverage, water availability and vapor pressure. The proposed work applied various machine learning techniques on the Indian crop production data set and analyzed these models based on various evaluation parameters for crop classification. The proposed work also implemented the deep learning technique and the results gives the comparative analysis of all the models to classify the crops. The results obtained after the experiments are promising and this helps the farmers to predict which crop to grow based on the weather and soil conditions.