Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Extending FKG.in: Towards a Food Claim Traceability Network

Created by
  • Haebom

Author

Saransh Kumar Gupta, Rizwan Gulzar Mir, Lipika Dey, Partha Pratim Das, Anirban Sen, Ramesh Jain

Outline

This paper proposes a Food Claims Traceability Network (FCN) that extends the Indian Food Knowledge Graph (FKG). The FCN presents an ontology design and semi-automated knowledge curation workflow, which were used to develop a proof-of-concept using Reddit data and a large-scale language model. The FCN integrates curated data inputs, a structured schema, and a provenance-aware pipeline for food-related claim extraction and verification. While directly linked to the Indian Food Knowledge Graph, the methodology presented here is application-agnostic and applicable to other regions, cuisines, and regulatory environments. By modeling food claims and their traceability in a structured, verifiable, and explainable manner, we aim to contribute to a more transparent and accountable food knowledge ecosystem, empowering researchers, policymakers, and consumers to navigate diverse dietary claims.

Takeaways, Limitations

Takeaways:
A new framework for traceability and verification of food-related claims (FCN) is presented.
Proposing an efficient knowledge curation method using Reddit data and LLM.
Presenting a methodology applicable to various regional and cultural contexts
Contributing to building a more transparent and responsible food knowledge ecosystem
Limitations:
Currently, it's based on a knowledge graph of Indian food. Expanding to other regions will require additional data and effort.
Consideration Needed for Bias and Reliability Issues in Reddit Data
Possibility of limited accuracy in assertion verification due to limitations of LLM
Further research is needed on the practical application and effectiveness of FCN.
👍