Milk Quality Prediction by Using a Supervised Machine Learning Algorithm and Arduino Uno

Adulteration of milk is common and observed recently with water and other substances like urea, whey, and some harmful substances in some cases. Existing instruments have some cons as they require procedures to be followed by skilled laborers and other related factors. Thus, the analysis of the quality of milk is performed to address this issue. In this study, the milk quality prediction is performed on applying Machine Learning algorithms. On comparing supervised machine learning algorithms like logistic regression, MP classifier, logistic regression, and random forest algorithm, it was concluded that XGBoost Classifier produced higher accuracy. Additionally, a connection is created between an arduino uno board that is given the instructions and connected with four sensors to measure the parameters and send them to the dataset through RS232 protocol. Then the values in the dataset are trained and tested using the machine learning algorithms resulting in a prediction thereby. This study overall helps with creating an effective and feasible outcome by reduction of manual processes. From milk factories to an ordinary milk consumer, this study provides an effective implication as it cuts down the complexity and requirement of skilled laborers by introducing instant prediction mechanisms in real time.