Spatial context-aware user mention behavior modeling for mentionee recommendation.

Area of Science:

• Social Media Analysis
• Information Science
• Human-Computer Interaction

Background:

• Mentions are key for user interaction and information spread on social media.
• Recommending mentionees (users to be mentioned) is crucial but under-researched for common users.
• Existing research often overlooks the contextual factors influencing mention behavior.

Purpose of the Study:

• To develop personalized mentionee recommendation techniques for social media.
• To model user mention behavior considering both semantic and spatial context.
• To enhance user interaction and information diffusion through accurate recommendations.

Main Methods:

• Analysis of real-world social media datasets to understand mention behavior.
• Development of the Spatial COntext-aware Mention behavior Model (SCOMM) for location-tagged mentioning activities.
• Design of an Item-Attribute Pruning (IAP) algorithm to optimize recommendation performance and handle large datasets.

Main Results:

• User mention behavior is significantly influenced by semantic and spatial context.
• SCOMM effectively models user preferences by integrating semantic and spatial factors.
• The proposed approach achieves superior effectiveness and efficiency in mentionee recommendations compared to state-of-the-art methods.

Conclusions:

• Spatial context is a vital factor in user mention behavior modeling.
• SCOMM provides a robust framework for personalized mentionee recommendations.
• The IAP algorithm enhances the scalability and performance of the recommendation system.