Representation Learning Techniques toConduct Odor Prediction

INST: L_042 ; This thesis work ensembles representational learning techniques for reducing the input data dimension and implementing classification algorithms for evaluating the performance of reduction techniques, namely, PCA, ICA, LDA, LLE, and t-SNE algorithms. The next step is to train transformed data with widely known classification algorithms such as Naïve Bayes, SVM, KNN, and Random Forest to discover a suitable method for the electronic nose.