Submit your paper 
Download Template JOURNAL DETAIL
A New Approach for Machine Learning-Based Recognition of Meat Species Using a BME688 Gas Sensors Matrix
Paper Type  |
Contributed Paper |
Title |
A New Approach for Machine Learning-Based Recognition of Meat Species Using a BME688 Gas Sensors Matrix |
Author |
Nursel SÖYLEMEZ MİLLİ, İsmail Hakkı PARLAK and Mehmet MİLLİ |
Email |
nurselsoylemez@ibu.edu.tr |
|
Abstract: Identifying meat species accurately is crucial for food safety, fraud prevention, and quality control in the food industry. Mislabeling or adulterating meat products can lead to economic losses and pose health risks to consumers. However, conventional methods for species identification, such as DNA analysis or spectroscopy, are often timeconsuming and expensive. In recent years, highly sensitive sensors have made estimating food product types and freshness possible. The BME688 sensor produced by Bosch Sensortec is one of the most sensitive gas measurement sensors today. In this study, various types of meat were classified using machine learning methods on the data obtained by the BME688 gas sensor. Each type of meat has a distinct microbiota composed of specific microorganisms that influence its spoilage process and the volatile compounds it releases over time. VOCs and VSCs released into the environment by microorganisms that develop over time in meat types can be detected with this sensor. In this study, Decision Tree, Gaussian Naive Bayes, Bagging Tree, Support Vector Machine, Xgboost, Logistic Regression, Multi-Layer Perceptron, And Bosch AI-Studio Neural Network models were trained and tested on data obtained from different meat types using BME688. The accuracy values of the trained models were compared, and it was determined that the GNB and BT models have the highest potential for possible usage scenarios. Considering the results obtained, it was revealed that the BME688 sensor could distinguish chicken, sheep, and cattle meat with near-perfect accuracy. |
|
Graphical Abstract: |
|
Article ID |
e2025031 |
Received Date |
2024-10-12 |
Revised Date |
2025-03-22 |
Accepted Date |
2025-04-21 |
Keyword |
BME688 gas sensor, odor recognition, meat identification, machine learning |
Volume |
Vol.52 No.3 In progress (May 2025). This issue is in progress but contains articles that are final and fully citable. |
DOI |
https://doi.org/10.12982/CMJS.2025.031 |
Citation |
MİLLİ N.S., PARLAK .H. and MİLLİ M., A New Approach for Machine Learning-Based Recognition of Meat Species Using a BME688 Gas Sensors Matrix, Chiang Mai Journal of Science, 2025; 52(3): e2025031. DOI 10.12982/CMJS.2025.031. |
SDGs |
|
| View:39 Download:0 | |
RELATED ARTICLE
Article ID: e2025017
Author:Qing Hu and Fenhua Zhu
Vol.52 No.2 (March 2025) View: 264 Download:107
Article ID: e2024095
Author:Zhen Liu, Yiwei Fu, Jiangsong Jiang, Ya Huang, Dong Li, Yikun Yue, Shaochun Yuan and Chengzhi Wang
Vol.51 No.6 (November 2024) View: 491 Download:231
page: 1 - 7
Author:Sitthichoke Subpaiboonkit[a], Chinae Thammarongtham[b] and Jeerayut Chaijaruwanich*[a,b,d]
Vol.39 No.1 (JANUARY 2012) View: 1,420 Download:3,015
Copyrights © Since 2021 All Rights Reserved by Chiang Mai Journal of Science