DiMB-RE introduces a powerful NLP framework for automatically extracting diet-microbiome associations from scientific literature at scale.

Key Innovations:

• Created a comprehensive annotated corpus with 15 entity types and 13 relation types specific to diet-microbiome interactions
• Developed specialized NLP models to identify these associations automatically
• Demonstrated potential for large-scale extraction of diet-microbiome knowledge

Business Impact: This technology enables personalized nutrition strategies based on gut microbiome profiles, supporting development of tailored dietary interventions for improved health outcomes and potential new product development in nutraceuticals and personalized nutrition.

DiMB-RE: Mining the Scientific Literature for Diet-Microbiome Associations