Crop yield prediction in big data environment

Abstract

Prediction of crop yield is one of the essential research areas which helps the farmers in selecting right crop and right pattern of cropping with respect to location and season.  The yield of crop is determined by weather conditions, soil characteristics and water characteristics of a location along with use of pesticides.  Prediction of yield is prerequisite for agricultural management.  Huge amount of agricultural data is being collected and archived as a routine activity by weather departments and agricultural departments.   The volume of data keeps on growing and big data technologies serves as an appropriate choice for storing as well as performing predictive tasks using machine learning algorithms.  In this research work, Hadoop Distributed File System (HDFS) based architecture is presented for crop yield prediction towards establishing a scalable architecture for prediction