Comparative Analysis on Diet Prediction of Diabetes Person Using Recurrent Neural Network Over Decision Tree with Improved Accuracy

The examination studies how efficient Recurrent Neural Networks and Decision Trees might improve diabetic meal recommendations. There are two groups in charge of implementation. The initial is a decision tree, and the second is an RNN. This execution used 1000 samples with 2000 samples. The G Power statistical tool generates samples with an 80 % probability. Present decision tree models are 66.31 % accurate, whereas recommended recurrent neural networks are <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 4. 4 3 \%}$</tex>. Recurrent neural networks outperform Bayes models. The significance level is 003. (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p_{i} 0.05$</tex>). The accuracy has improved and is now much greater than that of the decision tree.