With the rise of generative AI systems such as ChatGPT and Gemini, conversational agents are transforming travel planning and enabling seamless advertising recommendations. Yet personalization, transparency, and user trust remain challenging. This study proposes the Conversational Generative AI–Driven Advertising Recommendation Framework (CGAI-ARF), integrating large language models for intent understanding, generative models for persuasive ad content, and reinforcement learning for adaptive ranking. A hybrid engine combines dialogue history, user preferences, and contextual signals to deliver relevant and engaging ads. Experiments on real travel datasets show CGAI-ARF improves click-through rate by 10–18%, raises AUC and NDCG@10, and boosts conversion rate by 7% in user studies, enhancing satisfaction. Results highlight its potential for user-centric, ethical, and sustainable advertising in conversational AI travel applications.
Ho et al. (Mon,) studied this question.