Bottle Cleanliness Detection Device with Electronic Nose using PCA-KNN Algorithm

This paper proposes a compact Bottle Cleanliness Detection Device that uses the PCA-KNN algorithm and an Electronic Nose system to expedite bottle inspection procedures. This is especially important in areas like the Philippines, where bottle reuse is prevalent. Using machine learning-processed sensor data, the device uses smell analysis to distinguish between “clean” and “unclean” bottles. The investigation confirms the device’s reliability with a high accuracy rate of 94.67% and an F1 score of 94.52%. This research provides small and medium-sized businesses (SMEs) with a portable, affordable quality assurance solution supporting consumer safety.