Exploration of Data Fusion Strategies Using Principal Component Analysis and Multiple Factor Analysis

Abstract

In oenology, statistical analyses are used for descriptive purposes, mostly with separate sensory and chemistry data sets. Cases that combine them are mostly supervised, usually seeking to optimize discrimination, classification, or prediction power. Unsupervised methods are used as preliminary steps to achieving success in supervised models. However, there is potential for unsupervised methods to combine different data sets into comprehensive, information-rich models. This study detailed stepwise strategies for creating data fusion models using unsupervised techniques at different levels. Principal component analysis (PCA) and multiple factor analysis (MFA) were used to combine five data blocks (four chemistry and one sensory). The model efficiency and configurational similarity were evaluated using eigenvalues and regression vector (RV) coefficients, respectively. The MFA models were less efficient than PCA, having gradual distributions of eigenvalues across model dimensions. The MFA models were more representative than PCA, as indicated by high RV coefficients between MFA and each individual block. Therefore, MFA approaches were better suited for multi-modal data than PCA. This work approached data fusion systematically and showed the type of decisions that must be made and how to evaluate their consequences. Proper integration of data sets, instead of concatenation, is an important aspect to consider in multi-modal data fusion.

Approach strategies and application of metabolomics to biotechnology in plants

“…Oh et al 10.3389/fpls.2023.1192235 Frontiers in Plant Science frontiersin.org classification, and is also used as a method of resolving a problem called multicollinearity in regression analysis (Mafata et al, 2022).…”

“…PCA is mainly used when the purpose is to reduce dimensionality due to a large number of variables (as in the case with metabolomics). Principal Component Analysis (PCA) is used for sorting data into an order or classification, and is also used as a method of resolving a problem called multicollinearity in regression analysis ( Mafata et al., 2022 ). PCA is a method of reducing the dimensionality of the original variables by obtaining k (< n ) independent principal components from n variables, formed as a linear combination ( Chatfield, 2018 ) ( Figure 3 ).…”

Low-Cost Pocket Fluorometer and Chemometric Tools for Green and Rapid Screening of Deoxynivalenol in Durum Wheat Bran

“…Then, significant principal components coming from all models were merged into a predictor matrix. Data fusion at PCA level allows reduction in the level of noise introduced in the model, because only structural information from each source is retained [62,63].…”

Decoding the Identity of Pinot Gris and Pinot Noir Wines: A Comprehensive Chemometric Fusion of Sensory (from Dual Panel) and Chemical Analysis

“…Indeed, sensory evaluations are influenced by interactions between different sample components (e.g., matrix effects due to the chemical composition), and between these and the perception modes. In turn, the chemical composition can be linked to the viticultural–enological variables (e.g., grape variety, winemaking technique, vintage, and terroir, just to name a few); hence, sensory analysis results could at least, in theory, be described as a complex function of the chemical composition [ 8 , 9 ].…”