Author(s)

Jingru Chen

Corresponding Author

Jingru Chen

Abstract

In this study, we propose a comprehensive AI-driven framework to optimize regional marketing strategies in the restaurant industry, addressing the challenges of geographic taste variation, budget constraints, and campaign personalization. Our approach integrates transformer-based sentiment classification, LDA-based regional flavor modeling, and ROI optimization via predictive and economic models. A synthetic dataset of 10,000 simulated promotional campaigns across East, Southwest, and North regions was generated to evaluate three common strategies: event-based, health-driven, and influencer-led. Experimental results demonstrate that influencer campaigns achieve the highest ROI and conversion, particularly in urban regions. Sentiment scores show strong predictive power for campaign success, while LDA topic analysis reveals distinct regional flavor preferences. Our framework also includes A/B testing using LLM-generated copy and stochastic optimization under budget limits. The results suggest that AI-enabled personalization and regional adaptation can significantly improve promotional efficiency and customer engagement for both large chains and SMEs in the food service sector.

Keywords

Artificial Intelligence, Restaurant Marketing, Regional Adaptation, Sentiment Analysis, ROI Optimization