Prediction of Microbial Population in Meat Using Electronic Nose and Support Vector Regression Algorithm

Abstract

Meat is one of several sources of protein needed by the human body. Until now, meat consumption has continued to increase yearly for various reasons, including its high nutritional value as a source of protein and its wide availability. The public must know the excellent quality of meat not consume rotten meat when choosing meat. Meanwhile, people still use the sense of smell to determine the quality of meat based on personal views. To overcome these obstacles, it is necessary to develop a method to predict the microbial population in meat to determine whether the meat is safe for consumption. Prediction requires the use of an acceptable approach. Using an electronic nose (e-nose) in conjunction with the Support Vector Machine Regression (SVR) technique allows a structured approach to predict the microbial population in meat concerning the Meat Quality Standard. The prediction results indicate that the system is accurate, as shown by R2 0.977 and an RMSE 0.026