Title: Visual tools for 3-way analysis in R
Authors: Valentin Todorov - UNIDO (Austria) [presenting]
Michele Gallo - University of Naples Orientale (Italy)
Maria Anna Di Palma - L Orientale (Italy)
Abstract: The standard multivariate analysis addresses data sets represented as two dimensional matrices. In recent years, an increasing number of application areas like chemometrics, computer vision, econometrics and social network analysis involve analysis of data sets that are represented as multidimensional arrays and multi-way data analysis becomes popular as an exploratory analysis tool. The most popular trilinear models are PARAFAC and Tuccker3 and their results can be presented in several different ways, the first one being tables of the coefficients or loadings for each mode, either rotated or not. While it is important to inspect the numerical output of the methods for analysis of three-way data in order to properly interpret the results, of great help can be different visual representations of these outcomes. We present an R package, rrcov3way, implementing a set of functions for the analysis of three-way data sets, including PARAFAC and Tucker3 as well as their robust alternatives. Apart from basic tools for data handling and preprocessing of multidimensional arrays, tools for display of the raw data and the model results in two and three dimensional plots are provided.