The Unfolding Analysis for Symbolic Objects Based on the Example of the External Car Advertisement Evaluation

Authors

Keywords:

symbolic data analysis, unfolding analysis, preference measurement, car advertisements

Abstract

Aim: Multidimensional unfolding allows representing both columns (e.g. products, services) and rows (e.g. customers) of the preference matrix on the same low-dimensional map (usually it’s a two or three-dimensional map). The main aim of the paper was to propose how to perform unfolding analysis for symbolic objects. Methodology: The paper describes the possible ways of performing unfolding analysis for symbolic interval-valued data. The external unfolding is described in the details and used in the empirical part of the paper. The data (preferences and dissimilarities) were gathered by using the incomplete method of triads. Results: The empirical part presents an application for unfolding symbolic data to evaluate customers’ preferences, where car advertisements are used as the example. The results presented on a two-dimensional perceptual map allowed to discover seven groups of respondents with different preferences; most of them prefer Skoda, Audi, Volkswagen, and Honda advertisements to Toyota and Volvo. Implications and recommendations: The proposed external approach for symbolic data allows to represent objects as rectangles (on two-dimensional map) or cuboids (in the case of three dimensions). The respondents are represented as points. Further work should focus on creating an algorithm that allows for the presentation of both symbolic objects and preferences expressed by respondents in the form of rectangles or cuboids. Originality/Value: The paper presents an innovative and previously unpresented external unfolding for symbolic data. Besides that it presents how other unfolding approaches could be adapted for symbolic data.

Downloads

Download data is not yet available.

Author Biography

Marcin Pełka, Uniwersytet Ekonomiczny we Wrocławiu

Associate Professor, Faculty of Economics and Finance, Department of Econometrics and Computer Science

Downloads

Published

2024-03-06