대추형 방울토마토 품종의 전자코 기반 향기 패턴에 대한 감각특성 중심 해석 : 병렬 PCA, 보조변수 투영 및 MFA 접근의 비교

초록

This study examined how gas chromatography-based electronic-nose data could complement sensory interpretation of eight grape tomato cultivars characterized by generalized descriptive analysis (gDA). Aroma and flavor gDA data and electronic-nose data were analyzed separately using principal component analysis and agglomerative hierarchical clustering to compare cultivar differentiation structures. Electronic-nose variables were then projected as supplementary variables onto the gDA space, and multiple factor analysis integrated the two data blocks and evaluated their contributions. Significant cultivar differences were observed in sensory attributes and in mean peak areas of all electronic-nose compound classes. However, cultivar configurations and similarity structures from the two datasets were not closely aligned, consistent with an RV coefficient of 0.299. Supplementary projection placed electronic-nose variable groups along sensory-defined directions, whereas MFA showed dimension-dependent contributions of the two data blocks. Thus, electronic-nose data should complement, rather than validate or replace, sensory evaluation by providing volatile-pattern information for sensorydefined cultivar differences.

키워드