Topic Modeling and Link-Prediction for Material Property Discovery

2025-07-08

AI Summary

Large-scale, sparse, and noisy scientific literature networks often suffer from missing material–topic associations. To address this, we propose an AI-driven hierarchical link prediction framework that integrates hierarchical non-negative matrix factorization (HNMFk) with Boolean matrix factorization (BNMFk), augmented by logistic matrix factorization, to construct a three-layer interpretable topic tree. The framework automatically uncovers latent associations of transition metal dichalcogenides in domains such as superconductivity and energy storage. Leveraging probabilistic scoring and model self-selection, it achieves high-precision, traceable link prediction on a sparse knowledge graph comprising 46,000 publications. Experiments successfully recover masked superconducting material links and generate multiple interdisciplinary, empirically testable hypotheses. An accompanying interactive dashboard enables human-in-the-loop discovery.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Interdisciplinary science discovery with web data mining
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Link prediction infers missing or future relations between graph nodes, based on connection patterns. Scientific literature networks and knowledge graphs are typically large, sparse, and noisy, and often contain missing links between entities. We present an AI-driven hierarchical link prediction framework that integrates matrix factorization to infer hidden associations and steer discovery in complex material domains. Our method combines Hierarchical Nonnegative Matrix Factorization (HNMFk) and Boolean matrix factorization (BNMFk) with automatic model selection, as well as Logistic matrix factorization (LMF), we use to construct a three-level topic tree from a 46,862-document corpus focused on 73 transition-metal dichalcogenides (TMDs). These materials are studied in a variety of physics fields with many current and potential applications. An ensemble BNMFk + LMF approach fuses discrete interpretability with probabilistic scoring. The resulting HNMFk clusters map each material onto coherent topics like superconductivity, energy storage, and tribology. Also, missing or weakly connected links are highlight between topics and materials, suggesting novel hypotheses for cross-disciplinary exploration. We validate our method by removing publications about superconductivity in well-known superconductors, and show the model predicts associations with the superconducting TMD clusters. This shows the method finds hidden connections in a graph of material to latent topic associations built from scientific literature, especially useful when examining a diverse corpus of scientific documents covering the same class of phenomena or materials but originating from distinct communities and perspectives. The inferred links generating new hypotheses, produced by our method, are exposed through an interactive Streamlit dashboard, designed for human-in-the-loop scientific discovery.

Problem

• Predict missing links in large, sparse material science literature networks
• Discover hidden associations in transition-metal dichalcogenides (TMDs) research
• Enable cross-disciplinary hypothesis generation via topic-material link prediction

Innovation

• AI-driven hierarchical link prediction framework
• Combines HNMFk, BNMFk, and LMF techniques
• Interactive Streamlit dashboard for discovery