Analysing Model Performance: KNN and Random Forest-Enhanced KNN in Milk Quality Prediction

This research evaluates the effectiveness of two machine learning techniques in forecasting milk quality: the traditional K-Nearest Neighbors (KNN) algorithm and an enhanced hybrid model combining Random Forest with KNN (RF+KNN). Based on comprehensive evaluation metrics, the RF+KNN hybrid model demonstrates superior predictive capabilities with an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> value of 0.98, compared to 0.96 for the standard KNN model. The hybrid approach also achieves higher accuracy at 99.00% versus 98.00% for KNN, while maintaining a lower Mean Squared Error (MSE) of 0.01 compared to KNN's 0.02. These results indicate that integrating Random Forest's feature selection and weighting capabilities with KNN's classification approach creates a more robust milk quality prediction model. The findings suggest that the RF+KNN hybrid model offers significant advantages for quality control systems in the dairy industry, enabling more reliable and precise milk quality assessments than traditional methods.