.png){kind=icon}
Return
Deciphering leader decision-making patterns from open-source information: A temporal knowledge graph-based approach
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ipm.2026.105029.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In an era characterized by complex geopolitical and economic volatility, analyzing leader decision patterns is crucial for strategic foresight and organizational risk assessment. While leaders' decision-making footprints are increasingly captured in massive open-source information, deciphering these patterns presents significant challenges. Traditional decision-making analysis heavily relies on subjective expert judgment and struggles with large-scale unstructured data, whereas modern data-driven deep learning models often operate as uninterpretable black boxes lacking theoretical grounding. Furthermore, the inherent sparsity and temporal dispersion of open-source data complicate effective modeling. To address these limitations, this study proposes a theory-grounded computational framework that integrates a Temporal Knowledge Graph with a Decision Pattern Neural Network. The Temporal Knowledge Graph structures fragmented open-source information into temporally ordered semantic quadruples, enabling the dynamic representation of decision-related entities, relations, and contextual factors. The Decision Pattern Neural Network further models decision-making as a hierarchical process in which external environmental factors are filtered through leader-specific individual factors before producing action tendencies. In a case study of Saudi Aramco, we construct a Temporal Knowledge Graph containing 153,267 quadruples from more than 440,000 open-source records and evaluate the framework on 352 annotated decision events of CEO Amin Nasser. The proposed model achieves an overall prediction accuracy of 83.02%, with a precision of 86.49% and a recall of 88.89%. This study provides a theoretically sound and computationally scalable solution for deciphering leadership logic, effectively bridging the gap between qualitative decision theory and quantitative data analytics.
Keywords:
Decision patterns
Decision factors
Temporal knowledge graph
Open-source information
Journal
I
IF:
6.9
Papers:
578
Citations:
0
Organization
Cited Papers
No cited papers available