Please use this identifier to cite or link to this item:
https://digital.lib.ueh.edu.vn/handle/UEH/76225
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Lê Thị Hồng Minh | en_US |
dc.contributor.author | Nguyễn Bảo Trâm | en_US |
dc.contributor.other | Nguyễn Công Minh | en_US |
dc.contributor.other | Nguyễn Hữu Minh Duy | en_US |
dc.contributor.other | Võ Phước Khôi Nguyên | en_US |
dc.date.accessioned | 2025-08-29T03:38:11Z | - |
dc.date.available | 2025-08-29T03:38:11Z | - |
dc.date.issued | 2025 | - |
dc.identifier.uri | https://digital.lib.ueh.edu.vn/handle/UEH/76225 | - |
dc.description.abstract | Local markets are among the most iconic cultural attractions for visitors, where they can be fully immersed in experiencing Vietnamese daily life, traditional trading, and local cuisine while contributing to cultural preservation and economic development. With the increased volume of tourism-generated content through social media, there is greater potential for stakeholders in tourism to gauge the experiences and satisfaction of visitors to cultural attractions, such as local markets. This is quite impractical to handle manually; hence, the need arises for automated techniques of sentiment analysis to extract useful insights from such data. The given study will compare the performance of four sentiment analysis methods, LiuHu, VADER, SentiArt, and Multilingual, applied to tourist reviews in Ben Thanh Market, and determine which one is best for sentiment analysis. Qualitative interviews were also conducted to validate and provide further insight into the findings derived from sentiment analysis and topic modeling. A dataset comprising reviews from online tourism was assessed through the method of sentiment analysis. Methods used are based on the results derived from accuracy, recall, precision, and F1-score. In this study, geographical segmentation has been utilized to show variation in sentiments in regions. Latent topics from the reviews have been determined by applying topic modeling techniques. Tourists have been interviewed qualitatively to confirm and extend insights gained from quantitative findings. The results show that VADER outperforms other methods of sentiment analysis for the highest accuracy of sentiment classification. American tourists are found to be the most positive, and topic modeling shows the critical factors that lead to both satisfaction and dissatisfaction among tourists. Interviews provided a deeper understanding of the context and how cultural expectations and personal interactions shape the experiences of tourists. Results bring out concrete lessons for tourism operators and policymakers to drive forward market functions, ensuring experiences align with expectations from tourists. Strategies can thus be tailor-made to suit each business need by modifying the trend sentiments that can, therefore, effectively make up for marketing communication | en_US |
dc.format.medium | 84 p. | en_US |
dc.language.iso | en | en_US |
dc.publisher | University of Economics Ho Chi Minh City | en_US |
dc.relation.ispartofseries | Giải thưởng Nhà nghiên cứu trẻ UEH 2025 | en_US |
dc.subject | Experiential marketing | en_US |
dc.subject | Sentiment analysis | en_US |
dc.subject | Topic modeling | en_US |
dc.subject | Tourist satisfaction | en_US |
dc.subject | Cultural tourism | en_US |
dc.title | Bridging online and offline experiences: The role of experiential marketing in market tourism | en_US |
dc.type | Research Paper | en_US |
ueh.speciality | Khoa Kinh doanh quốc tế - Marketing | en_US |
ueh.award | Giải A | en_US |
item.openairetype | Research Paper | - |
item.openairecristype | http://purl.org/coar/resource_type/c_18cf | - |
item.languageiso639-1 | en | - |
item.cerifentitytype | Publications | - |
item.fulltext | Full texts | - |
item.grantfulltext | reserved | - |
Appears in Collections: | Nhà nghiên cứu trẻ UEH |
Files in This Item:
File
Description
Size
Format
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.