Intelligent System for Predicting Agricultural Drought for Maize Crop

Abstract

There has been little information in regard to agricultural drought prediction. This paper aimed at coming up with an efficient and intelligent agricultural drought prediction system.

By using a case study approach and knowledge discovery data mining process this study was preceded by drought literature review, followed by analysis of daily 1978-2008 meteorological and annual 1976-2006 maize produce data both from Voi, Taita-Taveta (Coast Province in Kenya).

The design and implementation of an agricultural drought prediction system, was made possible by computer science programming for meteorological data preprocessing, classification algorithms for training and testing as well as prediction and post processing of predictions to various agricultural drought aspects.

The study was evaluated by comparison of predicted with actual 2009 data as well as the Kenya Meteorological Department (KMD) 2009 records. The evaluation of this study results indicated consistency with the KMD 2009 outlook. The results showed that the application of classification algorithms on past meteorological data can lead to accurate predictions of future agricultural drought.

The recommendation is that future work can be based on designing a solution for multiple regions with multiple crops.