This work analyses brand positioning from online consumer reviews, combining text analysis with multivariate statistical techniques. The review texts are reduced to a small set of latent topics through the Latent Dirichlet Allocation model, and the estimated proportions are aggregated at the brand level. Multivariate analysis serves a twofold purpose, reducing dimensionality through Principal Component Analysis and grouping together brands that are described in similar terms through hierarchical clustering. The results are summarised in a perceptual map that provides a visual representation of the competitive positioning of brands as it emerges from the voice of consumers.
Il presente lavoro analizza il posizionamento dei brand a partire dalle recensioni online dei consumatori, combinando l'analisi testuale con tecniche di statistica multivariata. I testi delle recensioni vengono ricondotti a un insieme ridotto di temi latenti mediante il modello Latent Dirichlet Allocation, e le proporzioni stimate vengono aggregate a livello di brand. L'analisi multivariata viene impiegata con un duplice obiettivo, la riduzione della dimensionalità attraverso l'Analisi delle Componenti Principali e il raggruppamento dei brand descritti in modo simile tramite clustering gerarchico. I risultati vengono sintetizzati in una mappa percettiva che rappresenta visivamente il posizionamento competitivo dei brand così come emerge dalla voce dei consumatori.
Brand image e posizionamento competitivo: un'analisi multivariata delle recensioni online dei consumatori
VARGIU, BARBARA
2025/2026
Abstract
This work analyses brand positioning from online consumer reviews, combining text analysis with multivariate statistical techniques. The review texts are reduced to a small set of latent topics through the Latent Dirichlet Allocation model, and the estimated proportions are aggregated at the brand level. Multivariate analysis serves a twofold purpose, reducing dimensionality through Principal Component Analysis and grouping together brands that are described in similar terms through hierarchical clustering. The results are summarised in a perceptual map that provides a visual representation of the competitive positioning of brands as it emerges from the voice of consumers.| File | Dimensione | Formato | |
|---|---|---|---|
|
Vargiu_Barbara.pdf
Accesso riservato
Dimensione
1.03 MB
Formato
Adobe PDF
|
1.03 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/115429