Novel Approach for Forecasting Sugarcane Crop Yield: A Real-Time Prediction

Abstract

Agriculture is an environment in which there is considerable confusion. Crop development depends largely on several variables, including climate, temperature, genetics, politics and economics. In addition, a huge number of raw agriculture statistics are available, but study of those details for estimating crop yield is quite challenging. The most challenging job is therefore to include accurate details and awareness about the raw farm data. In order to evaluate cultivation yield, data mining could customize data expertise. The objective of this study was to predict crop yields through the use of data mining technological advances. Moreover, this paper compared various classification algorithms and it is expected that the results of the study may enhance the actual yields of sugarcane in a wide number of tropical fields. The specifications used in the forecast were plot (soil size, plant area, rain distance, previous year's plant yield), sugar-cane characteristics (cane class and sort), crop cultivation procedure (normal water resource size, cultivation technique, disease management process, sort / procedure of fertilizer) as well as rain quantity.