Decision-making for high-involvement products: Topic modelling using online reviews

Lee, Chanyoung (2020) Decision-making for high-involvement products: Topic modelling using online reviews. [Dissertation (University of Nottingham only)]

[img] PDF - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (999kB)


High-involvement products require a complex decision-making process. Automobiles require time and effort and a financial budget to purchase, requiring the consideration of diverse characteristics of the vehicle before purchase. Research in this area has focused primarily on finding the most significant features when purchasing an automobile using the traditional statistical method with surveys. However, advanced linguistic technique analysis provides an opportunity to extract meaning from the diverse comments provided by owners. In this paper, the author identifies the key topics of the customer decision-making process from electric automobile owners using a topic modelling approach with a latent Dirichlet allocation (LDA) model combined with natural language processing techniques. The dataset includes 956 free-text customer online reviews for Tesla. In an exploratory analysis involving electric automobiles, LDA uncovered 10 comprehensive lists of topics discussed by customers. Topics are key for electric automobile companies to manage their interactions with customers by understanding the primary interests and features in terms of the decision-making process of current and future customers. The proposed approach and findings are beneficial to support understanding customer perceptions. Through this method, the marketing and business strategy can be improved to maintain current customers and attract future customers.

Item Type: Dissertation (University of Nottingham only)
Depositing User: Lee, Chanyoung
Date Deposited: 14 Dec 2022 10:03
Last Modified: 14 Dec 2022 10:03

Actions (Archive Staff Only)

Edit View Edit View