Forecasting high-impact research topics via machine learning on evolving knowledge graphs

2024-02-13

arXiv.org

AI Summary

To address the challenge of early identification of scientific innovation, this paper proposes a forward-looking prediction framework for assessing the future academic impact of unpublished research directions. Methodologically, we construct a dynamic knowledge graph comprising 21 million scholarly papers, integrating semantic content and citation relationships—enabling, for the first time, quantitative prediction of impact for pre-publication scientific ideas. We introduce a joint modeling paradigm combining temporal graph neural networks, multimodal semantic embeddings, and dynamic link prediction. Our approach achieves an AUC exceeding 0.91, significantly outperforming baseline methods in identifying high-impact emerging topics at early stages. This work delivers an interpretable and scalable core predictive capability for AI-driven scientific discovery (“Artificial Muse”), thereby facilitating interdisciplinary innovation and optimizing the allocation of research resources.

Technology Category

• Data Mining & Knowledge Management: Semantic Web
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Computational Creativity

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: Interdisciplinary science discovery with web data mining

Abstract

The exponential growth in scientific publications poses a severe challenge for human researchers. It forces attention to more narrow sub-fields, which makes it challenging to discover new impactful research ideas and collaborations outside one's own field. While there are ways to predict a scientific paper's future citation counts, they need the research to be finished and the paper written, usually assessing impact long after the idea was conceived. Here we show how to predict the impact of onsets of ideas that have never been published by researchers. For that, we developed a large evolving knowledge graph built from more than 21 million scientific papers. It combines a semantic network created from the content of the papers and an impact network created from the historic citations of papers. Using machine learning, we can predict the dynamic of the evolving network into the future with high accuracy (AUC values beyond 0.9 for most experiments), and thereby the impact of new research directions. We envision that the ability to predict the impact of new ideas will be a crucial component of future artificial muses that can inspire new impactful and interesting scientific ideas.

Problem

• Predictive Analytics
• Research Impact
• Interdisciplinary Innovation

Innovation

• Machine Learning
• Knowledge Graph
• Predictive Impact Modeling