Aim: This research article aims to improve the accuracy rate in the Novel prediction of quality of rice in rice mills by using Random Forest (RF) in comparison with Decision Tree (D-Tree) Classifier. Materials & Methods: The data set in this paper utilizes the publicly available Kaggle data set for Novel prediction of the quality of rice in rice mills. The sample size of Novel prediction of quality of rice in rice mill with improved accuracy rate was sample 80 (Group 1=40 and Group 2 =40), and calculation is performed utilizing G-power 0.8 with alpha and beta qualities are 0.05, 0.2 with a confidence interval at 95%. Novel Prediction of quality of rice in rice mills with improved accuracy rate is performed by Random Forest (RF) whereas several samples (N=10) and Decision Tree (D-Tree) where the number of samples (N=10). Results: The Random Forest (RF) classifier has 94.0 higher accuracy rates when compared to the accuracy rate of Decision Tree (D-Tree) is 92.7. The study has a statistical significant value of p<0.05, i.e., p=0.0313. Conclusion: Random Forest (RF) provides better outcomes in accuracy rate when compared to Decision Tree (D-Tree) for Novel prediction of quality of rice in rice mills.