Application of Fine Decision Tree Machine Learning Algorithm to Predict the Subclinical Mastitis in Cow Milk Using Prototype E-nose

Abstract

AbstractIn the current scenario, conventional techniques such as California mastitis test (CMT), somatic cell count (SCC), and gas chromatography was used for the detection of subclinical mastitis affected in cow milk. The drawbacks in these methods were time-consuming and need to be trained people to use equipment, and equipments are not portable. In order to overcome the drawback of traditional methods, MOS sensor-based method has been proposed in this research work. The MOS sensor responses are faster, reliable, and accurate to detect the quality of the milk. The prototype E-nose has been developed to detect subclinical mastitis disease affected cow milk. The fine decision tree machine learning algorithm is also used to accurately predict the Mastitis disease affected milk. This study is an extension of our research work to develop the predictive model prototype E-nose using fine decision tree machine learning algorithm having Sensitivity 100% for cow milk with mastitis diseases and Specificity 98.03% correctly identifies cow milk without mastitis diseases and overall accuracy is 98.5%.KeywordsPrototype electronic noseSub-clinical mastitisFine decision tree