Modelling a user-centric interactive AI agent in healthcare applications

In our work we present a novel user-centric interactive AI agent, where we embedded advanced Natural Language Processing (NLP) techniques and GPT models to enhance personalized healthcare applications. The tool devel oped leverages demographics and historical interactions of the users in a novel system, which allows the agent to significantly improve context-awareness and personalize AI-generated responses. Using such approach we were able to develop a system which address the common under-use of both the capabilities of large language mod els and user data. The system integrates a Natural Language Understanding (NLU) module, a Memory-Structure User Profile, a Decision Module, and a Prompt Adjusting Mechanism, demonstrating a critical use case in conver sational AI for healthcare. We show a thorough experimental setup, including synthetic data generation, prompt comparison, with the primary evaluation, and fine-tuning of the language models with the subsequent agent simulations. Results indicated that enriched prompts incorporating contextual and user-specific information sig nificantly improved accuracy (up to 67.50%), coherence (up to 72.27%), personalization (up to 74.17%), and relevance (up to 68.33%) of responses across three distinct samples compared to prompts without additional information, with all improvements achieving statistical significance (𝑝 < 0.05).The results obtained align with the objectives of combining Natural Language Processing (NLP) applications with GPT transformers