The growing focus on healthy and sustainable eating requires innovative tools to support personalized recommendations. This talk will examine how Natural Language Processing (NLP) and Large Language Models (LLMs) can enhance food recommendation systems, enabling them to provide personalized suggestions that promote both individual well-being and environmental responsibility. In particular, we will first show the efectiveness of knowledge-aware recommendation models that encode information about healthy food consumption. Next, we emphasize the importance of natural language processing techniques, which can be used to nudge toward healthier food choices through automatically generated explanations. Finally, we will show recent advances aiming also to include the concept of sustainability in the design and development of conversational food recommenders. In particular, we will discuss a pipeline based on LLMs that identifies healthier and more sustainable food alternatives that can be suitable for the user. We will conclude the presentation by sketching several future directions of this exciting research line.
Bio