The semiconducting carbon nanotubes can be utilized for the fabrication of field-effect transistors. By measuring the transfer characteristics, carbon nanotube-based field-effect transistors (NTFET) could provide a large amount of useful information associated with different electrical/electronic properties. The interaction between the carbon nanotubes and target analyte may result in the changes in the NTFET characteristics, such as conductance, transconductance, and threshold voltage, which could be extracted as different features for the construction of training models. By applying different machine learning algorithms such as linear discriminant analysis (LDA), support-vector machine (SVM), k-nearest neighbors (KNN) and random forest, different species could be classified with the constructed model. We have developed different models for the classification of different targets including non-malignant/malignant cells (ACS Sens. 2017, 2, 1128-1132), live/dead cells (Anal. Chem. 2022, 94, 3565-3573), purine compounds (ACS Appl. Mater. Interfaces 2019, 11, 1219-1227), biogenic amines (Anal. Chem. 2022, 94, 3849-3857), tetrahydrocannabinol (ACS Sens. 2019, 4, 2084-2093), and mercury ions (Biosens. Bioelectron. 2021, 180, 113085). Presently, we are exploring the discrimination of norfentanyl in different biological fluids.