Gaussian process regression coupled with mRMR to predict adulterant concentration in cocaine

被引:0
作者
Anzanello, M. J. [1 ,4 ]
Fogliatto, F. S. [1 ]
John, D. [2 ]
Ferrao, M. F. [3 ]
Ortiz, R. S. [4 ,5 ]
Mariotti, K. C. [4 ,5 ]
机构
[1] Univ Fed Rio Grande do Sul, Dept Engn Prod & Transportes, Porto Alegre, RS, Brazil
[2] Univ Fed Rio Grande do Sul, Inst Quim, Programa Posgrad Quim, Porto Alegre, RS, Brazil
[3] Inst Nacl Ciencia & Tecnol Bioanalit INCT Bioanali, Campinas, SP, Brazil
[4] Superintendencia Policia Fed, NSCAD Microeletron, Porto Alegre, RS, Brazil
[5] Inst Nacl Ciencia & Tecnol Forense INCT Forense, Porto Alegre, Brazil
关键词
Cocaine adulterants; Wavelength selection; Gaussian Process regression; ReliefF; MRMR; ATR-FTIR; WAVE-NUMBER SELECTION; MAIN ADULTERANTS; CHARACTERIZE; SPECTROSCOPY; COUNTERFEIT; TOOL;
D O I
10.1016/j.jpba.2024.116294
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Street cocaine is often mixed with various substances that intensify its harmful effects. This paper proposes a framework to identify attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) intervals that best predict the concentration of adulterants in cocaine samples. Wavelengths are ranked according to their relevance through ReliefF and mRMR feature selection approaches, and an iterative process removes less relevant wavelengths based on the ranking suggested by each approach. Gaussian Process (GP) regression models are constructed after each wavelength removal and the prediction performance is evaluated using RMSE. The subset balancing a low RMSE value and a small percentage of retained wavelengths is chosen. The proposed framework was validated using a dataset consisting of 345 samples of cocaine with different amounts of levamisole, caffeine, phenacetin, and lidocaine. Averaged over the four adulterants, the GP regression coupled with the mRMR retained 1.07 % of the 662 original wavelengths, outperforming PLS and SVR regarding prediction performance.
引用
收藏
页数:7
相关论文
共 50 条
[21]   Parametric Gaussian process regression for big data [J].
Maziar Raissi ;
Hessam Babaee ;
George Em Karniadakis .
Computational Mechanics, 2019, 64 :409-416
[22]   Gaussian Process Regression for Fingerprinting based Localization [J].
Kumar, Sudhir ;
Hegde, Rajesh M. ;
Trigoni, Niki .
AD HOC NETWORKS, 2016, 51 :1-10
[23]   Gaussian process regression for tool wear prediction [J].
Kong, Dongdong ;
Chen, Yongjie ;
Li, Ning .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2018, 104 :556-574
[24]   RGBD Sensors Correction with Gaussian Process Regression [J].
Amamra, Abdenour ;
Aouf, Nabil .
2014 56TH INTERNATIONAL SYMPOSIUM ELMAR (ELMAR), 2014, :159-162
[25]   Parametric Gaussian process regression for big data [J].
Raissi, Maziar ;
Babaee, Hessam ;
Karniadakis, George Em .
COMPUTATIONAL MECHANICS, 2019, 64 (02) :409-416
[26]   Rectangularization of Gaussian process regression for optimization of hyperparameters [J].
Manzhos, Sergei ;
Ihara, Manabu .
MACHINE LEARNING WITH APPLICATIONS, 2023, 13
[27]   Humanoid environmental perception with Gaussian process regression [J].
Luo, Dingsheng ;
Ding, Yaoxiang ;
Han, Xiaoqiang ;
Ma, Yang ;
Deng, Yian ;
Liu, Zhan ;
Wu, Xihong .
INTERNATIONAL JOURNAL OF ADVANCED ROBOTIC SYSTEMS, 2016, 13 :1-14
[28]   Gaussian Process Based Sequential Regression Models [J].
Takegawa, Kaito ;
Yokoyama, Yuya ;
Hamasuna, Yukihiro .
2024 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN 2024, 2024,
[29]   Moderate deviations inequalities for Gaussian process regression [J].
Li, Jialin ;
Ryzhov, Ilya O. .
JOURNAL OF APPLIED PROBABILITY, 2024, 61 (01) :172-197
[30]   Gaussian Process Regression for Maximum Entropy Distribution [J].
Sadr, Mohsen ;
Torrilhon, Manuel ;
Gorji, M. Hossein .
JOURNAL OF COMPUTATIONAL PHYSICS, 2020, 418