WANG Ying-qi,ZHAO Han-qing,FANG Huan,WANG Tong.MALDI-TOF MS with Random Forest Fusion Model Applied to the Geographical Origin Traceability of Atractylodes Macrocephala Koidz.[J].Journal of Instrumental Analysis,2025,44(06):1147-1153.
WANG Ying-qi,ZHAO Han-qing,FANG Huan,WANG Tong.MALDI-TOF MS with Random Forest Fusion Model Applied to the Geographical Origin Traceability of Atractylodes Macrocephala Koidz.[J].Journal of Instrumental Analysis,2025,44(06):1147-1153. DOI: 10.12452/j.fxcsxb.240929422.
MALDI-TOF MS with Random Forest Fusion Model Applied to the Geographical Origin Traceability of Atractylodes Macrocephala Koidz.
In this study,the matrix-assisted laser desorption/ionization time of flight mass spectrometry(MALDI-TOF MS) was used to analyze the geographical origin traceability of Atractylodes macrocephala Koidz. in combination with two improved random forest fusion algorithms. Firstly,mass spectrum data of Atractylodes macrocephala Koidz. samples from 3 provinces were obtained by MALDI-TOF MS,and the data size of each sample was 1×234 154. In view of the large amount of each sample data,the data were preliminarily simplified by data bins strategy(1×6 600). Then,the principal component analysis was carried out to reduce the dimension by set the threshold of cumulative variance contribution rate. The dimensionality reduction data were used to construct the adaptive enhanced extreme random forest model(AERF) and the adaptive enhanced balanced random forest model(ABRF). Finally,AERF-ABRF model was obtained through model fusion strategy to trace the origin of Atractylodes macrocephala Koidz.. The results showed that the adaptive enhanced random forest model combined with model fusion strategy based on dimensionality reduction of data proposed in this study could accurately distinguish the samples from 3 provinces,and achieved correct classification rate(CCR) values of 100% for both the validation and test sample sets. At the same time,compared to individual models,the model fusion strategy exhibited a much higher correct classification rate.
关键词
Keywords
references
Guang H L , Qun Z , Su Q S , Kelvin S L , Hao Z , Zhong Z Z . J. Mol. Struct. , 2008 , 883 : 91 - 98 .
Wang Q , Yu Y J , Fu H Y , Li X , Yang J . J. Instrum. Anal. (王巧,于永杰,付海燕,李翔,杨健. 分析测试学报), 2023 , 42 ( 4 ): 389 - 401 .
Liu W , Qiu X W , Jiang L W , Fan W , Li P , Zheng Y . J. Instrum. Anal. (柳薇,邱熙文,蒋立文,范伟,李跑,郑郁. 分析测试学报), 2025 , 44 ( 2 ): 246 - 252 .
Zhang H , Lin C , Yin L , Si J , Yu M , Li J , Li L , Zhang T , Zou Z . Fitoterapia , 2023 , 165 : 105431 .
Yang S , Zhang J , Yan Y , Yang M , Li C , Zhong L , Gong Q , Yu H . Front. Pharmacol. , 2020 , 10 : 1629 .
Hu L , Chen X , Yang J , Guo L . Rapid Commun. Mass Spectrom. , 2019 , 33 ( 22 ): 1703 - 1710 .
Zhou J , Li Y , Zhao J , Xue X , Wu L , Chen F . Food Chem. , 2008 , 108 ( 2 ): 749 - 759 .
Gao L , Yang R W , Li L Q , Han B , Kai G Y . J. Appl. Res. Med. Aromat. Plants , 2025 , 46 : 100632 .
Pan Z , Xiong F , Chen Y L , Wan G G , Zhang Y , Chen Z W , Cao W F , Zhou G Y . Molecules , 2019 , 24 ( 24 ): 4478 .
Anastasaki E , Kanakis C , Pappas C , Maggi L , Campo C P , Carmona M , Alonso G L , Polissiou M G . Eur. Food Res. Technol. , 2009 , 229 : 899 - 905 .
Rapid Screening of 4 Banned Substances in Acne-clearing Cosmetics by UV Spectroscopy Combined with Machine Learning Algorithm
Research on Quantitative Analysis Method of Heavy Metal Elements in Soil Sediments Based on Laser Induced Breakdown Spectroscopy Combined with Machine Learning
Related Author
SHI Xin-hong
FANG Fang
LU Li
XIANG Jian-hua
ZHANG Tian-long
ZHOU Kui
WANG Ting
NIAN Juan-ni
Related Institution
Key Laboratory of Chemical Impurity Spectrometry,National Medical Products Administration
School of Traditional Chinese Medicine,China Pharmaceutical University
Jiangsu Institute for Drug Control
College of Chemistry Materials Science,Northwest University
College of Chemistry and Chemical Engineering,Xi'an Shiyou University