This study examines the effectiveness of combining Large Language Models
(LLMs) with traditional text mining methods to analyze Service Quality (SQ)
dimensions and customer satisfaction in both physical and omnichannel retail
settings. We investigate the usage of NLP techniques and LLMs to extract
sentiment and SQ dimensions from user-generated content (UGC). Analyzing
datasets of customer reviews (in English and Persian), we identify key factors
influencing customer satisfaction and dissatisfaction in physical retail and
omnichannel environments by applying unsupervised text mining to customer
reviews from supermarkets. Our analysis reveals that among the general SQ
dimensions, personal interaction, store policies, and product-related Dimensions
positively impact customer satisfaction, while reliability concerns contribute to
dissatisfaction. The importance of personal interaction is particularly pronounced
in smaller stores and towns. Conversely, hypermarkets should focus on improving
physical aspects and enhancing personal interaction to reduce negative feedback.
Integrating LLMs with text mining provides a comprehensive approach to
analyzing SQ dimensions across different retail formats, emphasizing the
necessity for ongoing human oversight to ensure the accuracy and reliability of
sentiment analysis and information extraction. Nonetheless, there are challenges,
such as discrepancies between model predictions and human judgments and
difficulties in accurately identifying specific dimensions from unstructured text.
| ISBN: | 978-80-7678-334-8 |
|---|
| EAN: | 9788076783348 |
|---|
| Počet stran |
46 stran |
| Datum vydání |
10. 06. 2025 |
| Pořadí vydání |
První |
| Jazyk |
anglický |
| Vazba |
e-kniha - pdf |
| Autor: |
Taha Falatouri |
| Nakladatelství |
Univerzita Tomáše Bati ve Zlíně |
| Tématická skupina |
999 - nezařazeno |
| Neprodejná publikace. Publikaci je možné poptávat zde: Volně dostupné na http://hdl.handle.net/10563/56858 |
|---|
Svaz českých knihkupců a nakladatelů