Predicting New Research Directions in Materials Science using Large Language Models and Concept Graphs

2025-06-20

AI Summary

To address the challenge of rapidly expanding materials science literature overwhelming researchers’ ability to track domain dynamics, this paper proposes an LLM-driven framework for前瞻性 research direction discovery. First, fine-tuned LLMs combined with prompt engineering precisely extract semantic concepts from large-scale abstracts. Second, a dynamically evolving concept graph is constructed, and temporal graph neural networks model cross-period conceptual relationships. Third, an interpretable and verifiable prediction evaluation mechanism is designed with expert feedback integration. This work pioneers the integration of LLM-based concept extraction with historical evolutionary modeling, overcoming limitations of traditional keyword statistics and manual reviews. Experiments demonstrate significant improvements in predicting emerging research combinations. In blind expert evaluation, over 70% of recommended directions were rated as both original and feasible, effectively supporting scientific ideation.

Technology Category

• Data Mining & Knowledge Management: Semantic Web
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

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

Abstract

Due to an exponential increase in published research articles, it is impossible for individual scientists to read all publications, even within their own research field. In this work, we investigate the use of large language models (LLMs) for the purpose of extracting the main concepts and semantic information from scientific abstracts in the domain of materials science to find links that were not noticed by humans and thus to suggest inspiring near/mid-term future research directions. We show that LLMs can extract concepts more efficiently than automated keyword extraction methods to build a concept graph as an abstraction of the scientific literature. A machine learning model is trained to predict emerging combinations of concepts, i.e. new research ideas, based on historical data. We demonstrate that integrating semantic concept information leads to an increased prediction performance. The applicability of our model is demonstrated in qualitative interviews with domain experts based on individualized model suggestions. We show that the model can inspire materials scientists in their creative thinking process by predicting innovative combinations of topics that have not yet been investigated.

Problem

• Predicting new research directions in materials science using LLMs
• Extracting concepts from scientific abstracts to find unnoticed links
• Training ML models to predict emerging concept combinations

Innovation

• LLMs extract concepts from scientific abstracts
• Concept graph built for literature abstraction
• ML predicts emerging concept combinations