Figure 2

This figure shows the application of logistic regression on the test set within the context of Linear Discriminant Analysis (LDA) when entire Raman spectra range (400–1750 cm⁻¹) is selected. (a) Raman spectra of 12 ferns, 12 monocots and 9 eudicots for both test and training sets. (b) The test set plot showcases the predictive modelling and evaluation of the test set using logistic regression, providing insights into the classification performance and accuracy of the LDA model. 5 ferns, 5 monocots, and 3 eudicots have been used for the test set. (c) The performance evaluation of the Linear Discriminant Analysis (LDA) model, showcasing accuracy, precision, recall determining its effectiveness in distinguishing different plant species. LDA successfully distinguishes among the three distinct plant types, with an accuracy of 91% in the test set and 100% accuracy in the training set. (d) The confusion matrix for the linear discriminant analysis (LDA) model provides a detailed breakdown of true positive, true negative, false positive, and false negative classifications, offering a complete assessment of the model's predictive performance for distinguishing plant species.