A Prognostic Reasoning Model for Improving Areacanut Crop Productivity using Data Analytics Approach

Abstract

A proactive decision-making process relies heavily on prognostic reasoning models. Due to the evolving agronomic conditions, prognostic reasoning models are now required in the agricultural sector for risk management and to increase the productivity of the most important plantation crops. The major goal of this study is to maximize areca nut crop productivity by identifying various combinations of the best features using the formal statistical test chi-square. By giving the questionnaires to the farmers growing the Arecanut crop in the Mangaluru area of Karnataka, the study’s real data set is created. The Nave Bayes, Random Forest, Logistic Regression, and Decision Tree classifiers are used to evaluate the best features discovered by the chi-square test. With a prediction accuracy of 99.67%, it has been discovered that the random forest outperforms other classifiers when it comes to crop yield.