Live Session
Thursday Posters
Research
Recommending Healthy and Sustainable Meals exploiting Food Retrieval and Large Language Models
Alessandro Petruzzelli (Dipartimento di Informatica - University of Bari), Cataldo Musto (Dipartimento di Informatica - University of Bari), Michele Ciro Di Carlo (Dipartimento di Informatica - University of Bari), Giovanni Tempesta (Dipartimento di Informatica - University of Bari) and Giovanni Semeraro (Dipartimento di Informatica - University of Bari)
Abstract
Many people are constantly seeking to make healthy food choices, but increasingly, we are also considering the impact our dietary habits have on the environment. This creates a complex challenge: how can we ensure that we are eating nutritious foods that nourish our bodies while also minimizing the environmental footprint of our meals?To address this issue, this paper proposes a novel framework called Healthy And Sustainable eating (HeASe). Given the rising global concerns about nutrition and environmental sustainability, individuals need effective tools to help them navigate these issues. HeASe leverages the latest advancements in artificial intelligence to empower users with knowledge and self-awareness.The framework works in two steps. First, it uses a food retrieval strategy that takes into account macro-nutrient information to identify alternative meals for a chosen recipe. This ensures that the substitutions maintain a similar nutritional profile. Next, HeASe employs large language models to re-rank these potential replacements while considering factors beyond just nutrition, such as the recipe's environmental impact and other user-defined preferences. The experimental phase of this research demonstrates the capabilities of LLM in identifying the more sustainable and healthy recipe within a set of candidate options. This highlights the potential of these models to guide users towards food choices that are both nutritious and environmentally responsible.