Search for a command to run...
Condiment sauces such as soy sauce, fish sauce, oyster sauce, and Worcestershire sauce play a vital role in culinary practices and cultural identity, particularly in the Philippines. These sauces are distinguished by their unique volatile organic compound profiles, which define their aroma and flavor. With the growing demand for these condiment products, there is an increasing need for accurate and efficient methods to classify them, ensuring product authenticity and strengthening quality control. However, conventional approaches such as sensory evaluation and laboratory-based chemical analysis are often expensive, time-consuming, and subjective. To address this limitation, we used an electronic nose (e-nose) system integrated with a Support Vector Machine (SVM) classifier for the classification of dark condiment sauces. The system consists of an array of MQ-series gas sensors connected to an Arduino Mega 2560 for analog-to-digital conversion, with Raspberry Pi 5 serving as the primary processing unit. Sensor data undergo preprocessing steps, including standardization and dimensionality reduction through principal component analysis, before being classified using SVM. A total of 120 samples, consisting of 40 readings per condiment type, were used for training and testing, while 60 additional samples—15 per class—were reserved for validation. The e-nose system achieved a 95% classification performance, as evaluated using a confusion matrix and overall accuracy metrics. These results demonstrate the potential of the e-nose combined with SVM as a reliable tool for condiment classification. The system offers practical applications in quality control and product authentication. Future work may extend its capabilities toward spoilage detection, the integration of different gas sensors, and the classification of a wider variety of condiment sauces.