Science Hierarchography: Hierarchical Organization of Science Literature

2025-04-18

AI Summary

The explosive growth of scientific literature impedes interdisciplinary association discovery and domain activity assessment. To address this, we propose a lightweight, efficient multi-dimensional hierarchical knowledge organization paradigm. Our method innovatively integrates embedding-based clustering (FastText/BERT) with LLM-driven semantic prompting (GPT-4), enabling non-exclusive, multi-perspective, and cross-granularity scientific knowledge structuring—from macro-level disciplines to micro-level research topics—while dynamically modeling domain density and knowledge gaps. Leveraging multi-label hierarchical encoding and graph-structure optimization, we construct an interpretable and navigable hierarchical knowledge graph. Experiments demonstrate that our approach significantly improves the accuracy and efficiency of LLM agents in locating target papers, outperforming both pure LLM-based tree construction and conventional retrieval methods in trend detection, interdisciplinary exploration, and interpretability.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Scientific knowledge is growing rapidly, making it challenging to track progress and high-level conceptual links across broad disciplines. While existing tools like citation networks and search engines make it easy to access a few related papers, they fundamentally lack the flexible abstraction needed to represent the density of activity in various scientific subfields. We motivate SCIENCE HIERARCHOGRAPHY, the goal of organizing scientific literature into a high-quality hierarchical structure that allows for the categorization of scientific work across varying levels of abstraction, from very broad fields to very specific studies. Such a representation can provide insights into which fields are well-explored and which are under-explored. To achieve the goals of SCIENCE HIERARCHOGRAPHY, we develop a range of algorithms. Our primary approach combines fast embedding-based clustering with LLM-based prompting to balance the computational efficiency of embedding methods with the semantic precision offered by LLM prompting. We demonstrate that this approach offers the best trade-off between quality and speed compared to methods that heavily rely on LLM prompting, such as iterative tree construction with LLMs. To better reflect the interdisciplinary and multifaceted nature of research papers, our hierarchy captures multiple dimensions of categorization beyond simple topic labels. We evaluate the utility of our framework by assessing how effectively an LLM-based agent can locate target papers using the hierarchy. Results show that this structured approach enhances interpretability, supports trend discovery, and offers an alternative pathway for exploring scientific literature beyond traditional search methods. Code, data and demo: $href{https://github.com/JHU-CLSP/science-hierarchography}{https://github.com/JHU-CLSP/science-hierarchography}$

Problem

• Organize scientific literature into hierarchical structure
• Balance computational efficiency with semantic precision
• Enhance interpretability and trend discovery in science

Innovation

• Combines embedding clustering with LLM prompting
• Captures multiple categorization dimensions
• Enhances interpretability and trend discovery