Scholarlink

The study is devoted to the development of conceptual and mathematical models for optimizing the interaction of actors in interdisciplinary scientific and technical programs (ISTP) with external partners in the context of a limited time window of technological opportunities. The central tool proposed is the use of graph neural networks (GNN) for structuring complex knowledge networks and managing coordination flows. Using the example of an interdisciplinary program in the chemical engineering sphere, hypotheses on the regional unevenness of the development of innovative technologies and the relationship between factors of the entrepreneurial environment and the dynamics of the innovation market are verified. A comprehensive methodology for assessing synergistic effects and constructing multi-level motivational models based on structural identifiers and classifiers integrated into the GNN architecture has been developed.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Modeling the Effect of Structuring and Managing Interdisciplinary Knowledge Based on Graph Neural Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • A. V. Roschyn,
  • B. I. Savelev

摘要

The study is devoted to the development of conceptual and mathematical models for optimizing the interaction of actors in interdisciplinary scientific and technical programs (ISTP) with external partners in the context of a limited time window of technological opportunities. The central tool proposed is the use of graph neural networks (GNN) for structuring complex knowledge networks and managing coordination flows. Using the example of an interdisciplinary program in the chemical engineering sphere, hypotheses on the regional unevenness of the development of innovative technologies and the relationship between factors of the entrepreneurial environment and the dynamics of the innovation market are verified. A comprehensive methodology for assessing synergistic effects and constructing multi-level motivational models based on structural identifiers and classifiers integrated into the GNN architecture has been developed.