Improving Classification Accuracy in Milk Quality Prediction through Advanced Machine Learning Techniques

View at publisher Summarise with VIRAT
Authors: Rajshree Soni, Vinod Maan, Prateek Bhanti
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

For food safety and the viability of the dairy sector, assessing milk quality is essential. However, current prediction models frequently have poorer accuracy and generalization due to class imbalance, a lack of data, and poorly tuned parameters. Using a dataset comprising important physicochemical and sensory characteristics, such as temperature, pH, taste, fat, odor, turbidity, and color, and with milk grade as target variable. The proposed work represents an comparative evaluation of the distinct machine learning models—Decision Tree, Random Forest, Support Vector Machine(SVM), Logistic Regression Model(LR), and Gradient Boosting—for the classification processes. The proposed research examines model performance throughout three stages: baseline(base-model), data augmentation, and hyperparameter tuning. At the starting stage the ensembled model such as Gradient Boosting and Random Forest acquired better outcomes with accuracies around 0.90. All models significantly improved after data augmentation, with Random Forest achieving an accuracy of 0.9974. Gradient Boosting achieved an accuracy of 0.9978 and an F1-score of 0.99 after additional improvement through hyperparameter tuning generated nearly perfect outcomes. With an accuracy of 0.9968, an ensembling of Random Forest, Gradient Boosting, and logistic Regression further increased reliability. Overall the results show that model ensembling, hyperparameter adjustment, and data augmentation significantly increase predicted accuracy and resilience, making ensemble approaches the preferred strategy for classification models.

Fields of Study & Subjects

Data Mining and Machine Learning Applications Smart Agriculture and AI Spectroscopy and Chemometric Analyses

Publication details

Journal / source
International Conference on Image Processing and Capsule Networks (ICIPCN)
Pages
1276–1281
Publication date
27 January 2026
Language
EN