基于引文网络结构变换的大数据研究前沿预测

Frontier Prediction of Big Data Based on Citation Network Structure Transformation

【目的/意义】对大数据知识领域的研究前沿及未来发展趋势进行预测。【方法/过程】利用引文网络结构变换模型方法,通过CiteSpace信息可视化软件工具,对下载于Web of Science检索平台的大数据研究领域文献进行分析,绘制文献共被引和引文结构变换知识图谱,分别从共被引文献和施引文献的视角,对大数据领域的研究前沿和未来发展趋势进行预测。【结果/结论】预测出了对学科发展具有潜在影响力、交叉属性最强、对中心度影响最大的文献以及相关研究主题。这种分析方法,避免了从施引文献或被引文献的单一角度分析知识领域研究前沿的不足,对系统地分析知识领域的前沿,预测未来潜在变化趋势具有一定的参考价值。

【Purpose/significance】This research presents a method for predicting the research fronts and future development within a scientific domain.【Method/process】 It employs citation structural variation models and puts forward a method to analyze scientific research fronts and make future predictions. It takes big data as an example and predicts its research fronts and future development from the perspective of both co-citation documents and citing documents.【Result/conclusion】The research predicts documents and relevant research topics with potential influence on the development of disciplines, the strongest interconnectivity, and the biggest influence on centrality. This research also has some limitations in data selection and fails to introduce the influence of relevant policies on this domain. The research avoids analyzing the research fronts in scientific domain from one perspective of citing documents or cited documents and has great reference value for systematically analyzing fronts in the scientific domain and predicting future trends.