基于SPO语义网络的核心主题识别与演化趋势分析方法

目的：应用自然语言处理技术和深层语义信息进行核心主题识别及演化趋势分析，帮助科研人员了解领域研究现状、跟踪领域研究热点、把握领域发展规律，进而推动学科领域的发展。方法：提出基于SPO语义网络的核心主题识别及演化趋势分析方法，首先抽取科技论文数据中的SPO结构，然后分阶段构建SPO语义网络，最后利用节点度和边权重等社会网络分析指标对领域核心主题及其演化趋势进行研究和探索。结果：选择基因编辑领域进行实证分析，识别出该领域的7个核心研究主题，并探究各主题内容及受关注程度的发展变化情况。结论：基于SPO语义网络的核心主题识别及演化趋势分析方法具有可行性和可靠性，可以为学科领域科研活动的展开提供重要决策支持。

Core theme identification and evolutionary trend analysis based on SPO semantic network

Objective The combined application of natural language processing and deep semantic information to identify core topics and analyze their evolutionary trends is helpful for researchers to understand the research status, track the research hotspots and grasp the law of development, and thus promote the development of the field. Methods A core theme identification and evolutionary trend analysis method was proposed based on SPO semantic network: firstly extracting the SPO structure of scientific and technical articles, then constructing SPO semantic network in stages, and finally using social network analysis metrics such as the degree of node and weight of edge to explore its core theme and evolutionary trend in this field. Results The field of gene editing was selected for empirical analysis, seven core research topics in this field were identified, and the developmental changes of the content and attention of each topic were explored. Conclusion The method is feasible and reliable, which provides important decision support for the development of scientific research activities in this field.