基于知识图谱自动构建的可用资源表示与推荐

Available Resources Representation and Recommendation Based on Automatic Construction of Knowledge Graphs

为了解决技术创新中可用资源选择过程算法化程度与创新效率较低的问题,提出基于知识图谱自动构建的可用资源表示与推荐方法;首先采用双向长短期记忆模型和条件随机场模型建立可用资源实体抽取模型,基于依存句法分析实现实体关系抽取,从而构建可用资源知识图谱;然后分析可用资源属性在可用资源利用过程中的作用,并架构属性匹配网络,实现基于知识图谱的可用资源推荐;最后通过实验和案例分析验证所提方法的有效性。结果表明,知识图谱能够有效表示可用资源,可以应用于智能化计算机辅助创新系统的开发。

To solve the problem of low algorithmic degree and innovation efficiency in the process of available resources selecting for technological innovation, an available resources representation and recommendation method based on automatic construction of knowledge graphs was proposed. An entity extraction model of available resources was firstly constructed by using bi-directional long short term memory model and conditional random field model. The relation extraction between entities was realized based on dependency parsing, and the knowledge graphs of available resources were constructed. Then the attribute mapping network was architected based on the mechanism analysis of available resources attributes in the process of available resources application to realize the available resources recommendation based on knowledge graphs. Finally, the effectiveness of the proposed method was verified by using experiments and case analysis. The results show that the knowledge graphs are efficient in available resources representation and can be applied to the development of intelligent computer aided innovation system.