<![CDATA[Accurate classification of aromatic rice types is crucial for maintaining quality and meeting consumer preferences. The purpose of this study is to classify MentikWangi, PandanWangi, and C4 rice based on their volatile content using e-nose. C4 rice, as a popular non-aromatic variety, was included as a comparison for sensor response analysis. The research method involved preprocessing the e-nose gas sensor readings, including feature extraction, baseline manipulation, and missing value checking, to ensure data quality. The classification was performed using Random Forest, Support Vector Machine, and Passive-Aggressive methods. The results showed that the Random Forest model achieved the highest accuracy of 97%, followed by the Support Vector Machine at 95% and Passive Aggressive at 90%. The model evaluation utilized a Confusion Matrix and Receiver Operating Characteristics, which confirmed that Random Forest was the best-performing model. This study concludes that e-nose-based classification effectively differentiates between aromatic rice types, providing significant potential for objective and efficient quality assessment and offering valuable insights for further research in areas such as food technology, agricultural science, and chemical analysis.]]>
Copyrights © 2025
