Comparison of Geographical Traceability of Wild and Cultivated Macrohyporia cocos with Different Data Fusion Approaches

Poria originated from the dried sclerotium of Macrohyporia cocos is an edible traditional Chinese medicine with high economic value. Due to significant difference in quality between wild and cultivated M. cocos, this study aimed trace origin fungus perspectives cultivation. In addition, there were quite limited studies about data fusion, a potential strategy, employed discussed geographical traceability cocos. Therefore, we traced cultivation using multiple fusion approaches. Supervised pattern recognition techniques, like partial least squares discriminant analysis (PLS-DA) random forest, using. Five types involving low-, mid-, high-level strategies performed. Two feature extraction approaches including selecting variables by forest-based method—Boruta algorithm producing principal components dimension reduction technique component analysis—were considered fusion. The results indicate following: (1) samples did exist terms content vital chemical fingerprint analysis. (2) Wild need realize traceability, accuracy validation set was 95.24%. (3) Boruta outperformed (PCA) extraction. (4) mid-level PLS-DA model took full advantage information synergy showed best performance. This proved that both optimal identification methods different, identification.