BibTex Citation Data :
@article{JPA33779, author = {Budi Sumanto and Ummi Kaltsum and Galih Setyawan and Ganjar Alfian and Dzulkifli Syauqi and Ariesta Handayani and Kombo Kombo}, title = {Environmental Compensation for Robust Tea Aroma Classification Based on an Electronic Nose and Machine Learning}, journal = {Journal of Physics and Its Applications}, volume = {8}, number = {3}, year = {2026}, keywords = {Electronic nose; MOS sensor; Tea aroma classification; Environmental compensation; machine learning}, abstract = { Classifying tea aromas using an electronic nose (e-nose) system offers rapid, non-destructive quality assessment. However, metal oxide semiconductor (MOS)-based gas sensors are often affected by temperature and humidity, reducing classifier robustness. To address this, we propose an environmental compensation approach to boost robustness in machine learning-based tea aroma classification. Specifically, we analyzed 400 tea samples (100 per class: black, green, red, yellow) using an e-nose with 10 MOS sensors under three scenarios: (A) sensor features only, (B) integration of temperature–humidity features, and (C) temperature–humidity-based signal compensation before feature extraction. For classification, we used SVM, Random Forest, KNN, and a soft voting ensemble. Notably, Scenario C performed best, achieving 85.00% accuracy with SVM. Furthermore, robustness analysis revealed that KNN led on test data (RI=0.9851), while SVM was perfect in cross-validation (RI=1.000). These results confirm that environmental compensation effectively improves the MOS e-nose system stability. }, issn = {2622-5956}, doi = {10.14710/jpa.v8i3.33779}, url = {https://ejournal2.undip.ac.id/index.php/jpa/article/view/33779} }
Refworks Citation Data :
Classifying tea aromas using an electronic nose (e-nose) system offers rapid, non-destructive quality assessment. However, metal oxide semiconductor (MOS)-based gas sensors are often affected by temperature and humidity, reducing classifier robustness. To address this, we propose an environmental compensation approach to boost robustness in machine learning-based tea aroma classification. Specifically, we analyzed 400 tea samples (100 per class: black, green, red, yellow) using an e-nose with 10 MOS sensors under three scenarios: (A) sensor features only, (B) integration of temperature–humidity features, and (C) temperature–humidity-based signal compensation before feature extraction. For classification, we used SVM, Random Forest, KNN, and a soft voting ensemble. Notably, Scenario C performed best, achieving 85.00% accuracy with SVM. Furthermore, robustness analysis revealed that KNN led on test data (RI=0.9851), while SVM was perfect in cross-validation (RI=1.000). These results confirm that environmental compensation effectively improves the MOS e-nose system stability.
Article Metrics:
Last update:
As an article writer, the author has the right to use their articles for various purposes, including use by institutions that employ authors or institutions that provide funding for research. Author rights are granted without special permission.
Author who publishes a paper at Journal of Physics and Its Applications (JPA) has the broad right to use their work for teaching and scientific purposes without the need to ask permission, including: used for (i) teaching in the author's class or institution, (ii) presentation at meetings or conferences and distributing copies to participants ; (iii) training conducted by the author or author's institution; (iv) distribution to colleagues for research use; (v) use in the compilation of subsequent authors' works; (vi) inclusion in a thesis or dissertation; (vi) reuse of part of the article in another work (with citation); (vii) preparation of derivative works (with citation); (viii) voluntary posting on open websites operated by authors or author institutions for scientific purposes (follow the CC BY-SA License).
Authors and readers can copy and redistribute material in any media or format, and mix, modify, and build material for any purpose but they must provide appropriate credit (provide article citation or content), providing links to the license, and indicate if there are changes.
The authors submitting a manuscript do so on the understanding that if accepted for publication, copyright of the article shall be assigned to Journal of Physics and Its Applications (JPA). Copyright encompasses rights to reproduce and deliver the article in all form and media, including reprints, photographs, microfilms and any other similar reproductions, as well as translations.
Reproduce any part of this journal, its storage in the database or its transmission by all forms or media is permitted does not need for written permission from JPA. However, it should be cited as an honor in academic manners
JPA and the Department of Physics Diponegoro University and the Editor make every effort to ensure that there are no data, opinions, or false or misleading statements published in JPA. However, the content of the article is the sole and exclusive responsibility of each author.
The Copyright Transfer Form can be downloaded here: [Copyright Transfer Form JPA]. The copyright form should be signed originally and send to the Editor in the form of printed letters, scanned documents sent via email or fax.
Dr. Eng. Ali Khumaeni, M.E. (Editor in Chief)
Editorial Office of Journal of Physics and Its Applications (JPA)
Department of Physics, Faculty of Sciences and Mathematics, Diponegoro University
Journal of Physics and Its Applications (JPA) (e-ISSN: 2622-5956) is published by the Department of Physics, Faculty of Science and Mathematics, Diponegoro University, Indonesia under Creative Commons Attribution-ShareAlike 4.0 International License.
