ABSTRACT
Flavoromics studies the chemical compound profiles of flavor (taste and aroma) using a data-driven methodology to correlate chemical compound profiles with the sensory properties of foods. Data mining is meant to extract meaningful knowledge from useful but non-evident information hidden within large datasets. It is performed to extract meaningful information from the samples and visualize the results. The learning of principles of common supervised and unsupervised multivariate statistical tools is important to select the right data mining methods. Therefore, this chapter introduces techniques for data mining and data analysis tasks, ranging from classification analysis, regression analysis, correlation analysis, and cluster analysis. For each topic, it covers basic concepts, task formulations, methodologies, and evaluation metrics.
