期刊论文

Xiang, L.

中文

数据分析与知识发现

2096-3467

2023

7

2

119-128

10.11925/infotech.2096-3467.2022.0330

The work is supported by National Natural Science Foundation of China (Grant No. 71673106).

本校为第一机构

【目的】通过专利数量和专利引用识别明星发明人类型的方法存在明显时滞效应，本文结合专利文本和发明者合作关系构建了图卷积神经网络，该模型可以用于明星发明人的早期识别。【方法】从“延续性创新”、“突破性创新”两个维度将明星发明人的创新类型分为“复合型”、“巩固型”、“突破型”和“发展型”4类，结合专利标题信息和明星发明人的合作关系，构建基于图卷积神经网络的明星发明人类型的早期识别模型。【结果】以分子生物学与微生物学领域内专利数据进行了验证，实验表明本模型识别明星发明人创新类型的整体准确率为79.4%，相较于只使用词向量的方法准确率提高了约15个百分点。【局限...

[Objective]Identifying the star inventors by the number of patents and patent citations has obvious time lag effects.Therefore,this paper constructs a graph convolutional neural network to find the emerging star inventors effectively.[Methods]This paper defines four typ...MORE [Objective]Identifying the star inventors by the number of patents and patent citations has obvious time lag effects.Therefore,this paper constructs a graph convolutional neural network to find the emerging star inventors effectively.[Methods]This paper defines four types...