EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

The application of deep learning in economic analysis and marketing strategy formulation in the tourism industry

Jing Zhang, Ming Gao iD

PLOS One · 2025

DOI: 10.1371/journal.pone.0321992 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Jing Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: Hospitality Management Department, Tourism College of Zhejiang, Hangzhou, China. Hauptthemen laut MEDLINE: Marketing, Deep Learning, Tourism. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Social Science Foundation Art Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The tourism industry is ever-evolving in nature, as it operates in a global marketplace that has become progressively global and offers great potential due to technological advances. The tourism industry faces challenges in accurately forecasting economic impacts and understanding visitor patterns that rapid global changes. Motivated by these needs, this research introduces the Tourism Variational Recurrent Neural Network (TourVaRNN), aiming to enhance the tourism industry by predicting economic impacts and visitor behaviors for effective marketing strategies through advanced Deep Learning (DL) techniques. The research applies a variational recurrent neural network for enhancing tourism demands and model the complex temporal dependencies within tourism data. The proposed TourVaRNN integrates variational autoencoders to capture latent variables representing visitor preferences and spending habits, while recurrent neural networks model complex temporal dependencies in tourism data. Marketing campaigns in the tourism sector can be fine-tuned through visitor segmentation, which seeks to comprehend and classify visitors according to their demographics, preferences, and behaviors. The model employs robust forecasting of economic impacts, visitor spending patterns, and behavior while accounting for uncertainty through variational inference. The implementation uses Python language on a tourism dataset comprising necessary attributes like visitor numbers, days, spending patterns, employment, international tourism samples over a specific region, and a diverse age group analyzed over a year. The proposed method is evaluated in terms of performance metrics such as economic impact assessment, visitor segmentation efficiency, inference time analysis, and budget allocation utilization for effective economic and marketing strategy analysis in the tourism industry. TourVaRNN’s improved segmentation efficiency of 15.7 percent allows for more targeted marketing, increasing engagement with visitors and income. Decisions may be made in real-time, improving operational efficiency in tourism management, thanks to a 17.5% reduction in inference time (to 40 ms). The most efficient use of funds is guaranteed by a 13.4% rise in budget allocation utilization, leading to maximum economic benefits.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MarketingDeep LearningTourismHumansModels, EconomicNeural Networks, Computer

Autor:innen und Institutionen

  • Hospitality Management Department, Tourism College of Zhejiang, Hangzhou, China.

ORCID-Profile: Ming Gao

Förderung

  • National Social Science Foundation Art Project

Zitieren

APA

Zhang, J., & Gao, M. (2025). The application of deep learning in economic analysis and marketing strategy formulation in the tourism industry. PLOS One. https://doi.org/10.1371/journal.pone.0321992

Vancouver

Zhang J, Gao M. The application of deep learning in economic analysis and marketing strategy formulation in the tourism industry. PLOS One. 2025. doi:10.1371/journal.pone.0321992

BibTeX
@article{Zhang2025,
  author = {Zhang, Jing and Gao, Ming},
  title = {The application of deep learning in economic analysis and marketing strategy formulation in the tourism industry},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0321992}
}
Kostenlos starten →