Enabling personalized smart tourism with location-based social networks

Area of Science:

• Computer Science
• Information Science
• Tourism Studies

Background:

• The tourism industry is rapidly evolving due to advancements in mobile internet, IoT, and communication technologies, leading to the rise of smart tourism.
• Location-based social networks (LBSNs) offer rich data for smart tourism but face challenges in multi-source information modeling and data sparsity.
• Personalized travel planning and recommendations are key features of smart tourism, requiring effective data analysis techniques.

Purpose of the Study:

• To propose a novel knowledge-driven personalized recommendation method for smart tourism by leveraging deep learning on LBSN data.
• To address the complexities of multi-source information modeling and data sparsity inherent in LBSN data for tourism applications.
• To enhance user travel planning and recommendations by exploring personalized travel behaviors and understanding human mobility patterns.

Main Methods:

• Utilized representation learning techniques to model contextual information (time, space, semantics) within LBSNs.
• Employed contrastive learning techniques for data augmentation to explore user behaviors and mitigate data sparsity.
• Conducted a case study focused on trip recommendation to validate the proposed approach.

Main Results:

• The proposed knowledge-driven method effectively models contextual information from LBSNs for personalized recommendations.
• Contrastive learning strategies successfully alleviated data sparsity and revealed user personalized travel behaviors.
• The case study demonstrated the practical effectiveness of the approach in trip recommendation.

Conclusions:

• The developed deep learning method offers a robust solution for personalized smart tourism recommendations using LBSN data.
• The approach effectively handles complex data challenges, paving the way for more intelligent tourism services.
• Insights into human mobility patterns were gained by analyzing contextual data and tourist preferences.