Crop Yield Prediction in Agriculture Using Gradient Boosting Algorithm Compared with Random Forest
MYS. Karthik Yasaswy, C. T. Manimegalai, Jayalakshmi Somasundaram
Abstract
To predict the crop yield rate using Innovative Gradient Boosting over Random Forest in Agriculture. Random Forest and Gradient Boosting with sample size (N=20) were iter-ated 10 times to predict Crop Yield accuracy. Novel Crop Yield Prediction includes predicting the yield of the crop from previous historical data. Results and The accuracy for Novel Crop Yield Prediction in Innovative Gradient Boosting (98.7 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> ) and Random Forest (86.4%) is obtained. There was a statistical significance between Gradient Boosting and Random Forest <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathbf{p}=\mathbf{0.005})$</tex> . For each algorithm, 20 samples iterated and G power were calculated as actual power of 80% and alpha value set as 0.025. Detection of Crop Yield using Innovative Gradient Boosting algorithm appears to be significantly better than Random Forest with improved accuracy.