Detection of ethanol quality using random forest algorithm in comparison with decision tree algorithm to measure accuracy.

This article's objective is to compare the Random Forest algorithm to the Decision Tree method to assess the quality of ethanol. The Wine Quality Reds dataset used in the proposed work is taken from the UCI library and has a total sample size of 1599. Training data (n=1199; 75% of total samples) and testing data (n=400; 25% of total samples) are separated from the gathered samples. Alpha and power are two distinct groups that are calculated using the G power tool. Random forest technique accuracy score values are computed for better ethanol quality identification. In comparison to the Decision Tree method, the accuracy of Random Forest was higher (82.21%). The model's derived significance value is 0.00 (p 0.05), and G power is discovered to be 0.8. This study discovered that the Random Forest algorithm detects ethanol quality substantially more accurately than the Decision Tree technique. [ABSTRACT FROM AUTHOR]