Discrimination between Alternative Herbal Medicines from Different Categories with the Electronic Nose

Abstract

As alternative herbal medicine gains soar in popularity around the world, it is necessary to apply a fast and convenient means for classifying and evaluating herbal medicines. In this work, an electronic nose system with seven classification algorithms is used to discriminate between 12 categories of herbal medicines. The results show that these herbal medicines can be successfully classified, with support vector machine (SVM) and linear discriminant analysis (LDA) outperforming other algorithms in terms of accuracy. When principal component analysis (PCA) is used to lower the number of dimensions, the time cost for classification can be reduced while the data is visualized. Afterwards, conformal predictions based on 1NN (1-Nearest Neighbor) and 3NN (3-Nearest Neighbor) (CP-1NN and CP-3NN) are introduced. CP-1NN and CP-3NN provide additional, yet significant and reliable, information by giving the confidence and credibility associated with each prediction without sacrificing of accuracy. This research provides insight into the construction of a herbal medicine flavor library and gives methods and reference for future works.

AI-based analysis of CT images for rapid triage of COVID-19 patients

“…Several feature engineering methods were applied: (1) SMOTEENN (synthetic minority oversampling technique and edited nearest neighbors) 48 : The method performs oversampling using SMOTE and cleaning using ENN to deal with imbalanced classes. In this study, a 1:1 (positive cases: negative cases) balanced dataset and a 1:3 imbalanced dataset were created respectively; (2) SMOTEENN + PCA (principal component analysis) [49][50][51] : upon enlarging the dataset, PCA was applied to reduce the dimensionality of the features. It applies singular value decomposition (SVD) to find the orthogonal principal components and the low-dimension representation of data.…”

Rapidly and accurately estimating brain strain and strain rate across head impact types with transfer learning and data fusion

“…First, we extracted the 160 temporal features (based on angular velocity and angular acceleration) that were used in our previously developed MLHMs [25], which included 20 features for each channel of signal: the maximum values, the minimum values, the integral values, the integral of absolute values, three sets of maximum and minimum values of the exponential moving average of the signal derivatives with different smoothing factors [38], [39], and the extrema information (the numbers of all positive extrema and of all negative extrema and the second to the fifth positive/negative extrema values). These features were extracted because they took both the signal intensities and time histories into consideration and proved to be effective in accurately predicting whole-brain strain, with specific details and reasons shown in our previous study [25].…”

Bio-Inspired Strategies for Improving the Selectivity and Sensitivity of Artificial Noses: A Review

“…To tackle the high dimensionality, it is crucial to select the optimal set of features to cover most of the information. Calculating the correlation between them is a classical method to do so [178,179]. Perera et al [180] chose a bioinspired technique, which allowed a robust classification, even with a small number of samples.…”