An LLM-based Hierarchical Framework for Constructing the Disciplinary Knowledge System of Neuromanagement and Neuroengineering

Abstract

The concept of the discipline "Neuromanagement" was first introduced by Chinese scholar in 2006, aiming to apply neuroscience and technological methods to study psychological and behavioral issues in management. Over the past two decades, both Neuromanagement and Neuroengineering have advanced significantly in theoretical exploration and practical applications, evolving into an international emerging discipline within management science and engineering. On the eve of the twentieth anniversary of this conceptual breakthrough, a comprehensive examination of the disciplinary evolution and accumulated knowledge is of milestone significance. The study proposed a novel large language model (LLM) based hierarchical framework for constructing the disciplinary knowledge system of Neuromanagement and Neuroengineering, from phenomenon discovery to causal revelation, mechanism exploration, and theory building. It centers on three core dimensions: knowledge systematization, methodological innovation, and multi-scenario application, and is organized into three layers: theoretical foundation, technological support, and application. By utilizing LLM to extract and hierarchically classify knowledge from extensive literature, the framework generates a structured knowledge graph. Empirically, the analysis synthesizes innovative outcomes generated across diverse hierarchical scenarios, thereby substantiating the practical value and guiding significance of the constructed knowledge system. This work provides an innovative theoretical schema and methodological approach for constructing disciplinary knowledge system of Neuromanagement and Neuroengineering, contributing to a reference paradigm for addressing the issue in diverse fields.