Decoding Plastic Toxicity: An Intelligent Framework for Conflict-Aware Relational Metapath Extraction from Scientific Abstracts

2025-09-14

AI Summary

Micro- and nanoplastics (MNPs) pervasively contaminate the environment, and multi-pathway human exposure has been linked to respiratory, gastrointestinal, and neurological disorders. However, existing studies lack systematic causal modeling across the “pollution source → exposure pathway → health effect” chain. To address this, we propose a novel framework integrating large language models (LLMs) with a dynamic evidence reconciliation mechanism for high-precision, robust relational meta-path extraction from scientific abstracts. Our method first identifies entities and cross-sentence relations, then performs multi-step semantic reasoning to generate structured toxicity propagation trajectory graphs. Crucially, we introduce a dynamic evidence reconciliation module that explicitly models and resolves semantic conflicts across heterogeneous texts. Evaluated on noisy, real-world literature data, our approach significantly improves both accuracy and interpretability of extracted causal paths. It enables automated, scalable mining and construction of complex causal knowledge graphs for MNP-related health risks.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data integration and cleaning
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The widespread use of plastics and their persistence in the environment have led to the accumulation of micro- and nano-plastics across air, water, and soil, posing serious health risks including respiratory, gastrointestinal, and neurological disorders. We propose a novel framework that leverages large language models to extract relational metapaths, multi-hop semantic chains linking pollutant sources to health impacts, from scientific abstracts. Our system identifies and connects entities across diverse contexts to construct structured relational metapaths, which are aggregated into a Toxicity Trajectory Graph that traces pollutant propagation through exposure routes and biological systems. Moreover, to ensure consistency and reliability, we incorporate a dynamic evidence reconciliation module that resolves semantic conflicts arising from evolving or contradictory research findings. Our approach demonstrates strong performance in extracting reliable, high-utility relational knowledge from noisy scientific text and offers a scalable solution for mining complex cause-effect structures in domain-specific corpora.

Problem

• Extracting multi-hop semantic chains linking pollutants to health impacts
• Resolving semantic conflicts from evolving or contradictory research findings
• Mining complex cause-effect structures in scientific domain corpora

Innovation

• LLM-based relational metapath extraction
• Toxicity Trajectory Graph construction
• Dynamic evidence reconciliation module