Who Really Influences? A Theory-Informed AI Framework for Tourism Influencer Selection Under Budget Constraints

Abstract

Social media influencers are particularly attractive to marketers because of their substantial followers and cost-effective influence. However, research is insufficient on selecting appropriate influencers to maximize the impact of tourism marketing within budget constraints. Drawing on complex contagion theory and the influence maximization problem, we develop a theory-informed AI framework to select social media influencers, maximizing the influence of tourism marketing. This approach consists of a diffusion model and an influencer inference algorithm. Extensive experiments on simulated and real datasets demonstrate that our approach outperforms existing approaches in inferring influencers, capturing diffusion patterns and scalability, and is applicable to tourism social media marketing. Our application of complex contagion theory extends theoretical understanding in tourism social network marketing. This approach works across various tourism contexts and scales, thus providing marketers accurate tools to identify influencers who generate stronger user engagement and marketing impact.