Random Forest Algorithm for Meat Classification and Microbial Population Prediction

Meat is one of the various protein sources that needed for human body. Now, the consumption of meat has always increased from year to year due to various factors, including having high nutrition, as a source of protein, to its distribution which can be found almost everywhere. In selecting meat, many consumers do not know about the quality of the meat offered by the seller, both the duration and the preservatives used by the seller in marketing the meat. The conventional way buyers use to determine the quality of meat is by checking the smell of the meat using the nose manually. In overcoming these problems, determining the quality of the meat needed the right method in checking. By applying the Random Forest Classification and Regression method, Electronic Nose can work structured on every component needed to determine meat quality. This experiment showed that the Random Forest Classification and Regression algorithm obtained that the best parameter for classification is n estimators 72 with result mean test score is 0.954955 and the best parameter for regression is n estimators 233 with result RMSE is 0.0141 and R2 is 0.9876.