Subjects: Library Science，Information Science >> Information Science submitted time 2017-11-08 Cooperative journals: 《数据分析与知识发现》
Abstract：【目的】通过构建共主题网络, 对主题之间的关系进行分析, 优化主题包含的词项。【方法】将“文档–主 题”二分图依照加权投影规则生成共主题网络, 使用介数中心性和主题概率结合的方法测度共主题网络中重点主 题, 通过 GN 算法对主题网络进行社区分割, 使用相关度方法优化主题词项。【结果】将共主题网络与基于 JSD 的 K-means 方法进行比较发现, 两者在三种主题数(最优主题数 28 和随机主观主题数 20, 30)测试下产生的聚类 数目都相同, 聚类内容的一致程度分别达到 100%、95%、87%。【局限】其他社区分割方法共主题网络未能全面涉 及。【结论】共主题网络照顾到了高维数据的需要, 能够探查出文档中哪些主题是重要主题, 哪些主题联系紧密。
Subjects: Library Science，Information Science >> Library Science submitted time 2024-01-10 Cooperative journals: 《文献与数据学报》
Abstract： ［Purpose/significance］Outlining and analyzing the scholarly landscape of the Pre-Qin thinkers and sorting out the relationship between them will be of great help in discovering their inheritance and influence.［Method/process］Using text reuse as a tool, the paper first builds a citation network to analyze the citations of the texts of the thinkers; then, examines the citations of the Analects by Mencius and Xunzi, and analyzes the ideas that they have absorbed and transformed from the Analects; and lastly, examines who is more in line with the Synthesizer between the Lüshi Chunqiu and the Xunzi. ［Result/conclusion］Hanfeizi and Xunzi is more connected with other scholars in the citation network, which indicates the realist position of Pre-Qin thought. In the citations to the Analects, Mencius focuses on the promotion of virtue and Xunzi on the reflection of realist character in spirit. The citation network reveals that the Lüshi Chunqiu is more in line with Synthesizer than the Xunzi.
Subjects: Library Science，Information Science >> Readers Work submitted time 2023-11-07 Cooperative journals: 《文献与数据学报》
Abstract： ［Purpose/significance］As one of the earliest news pictorial in China, Dianshizhai Pictorial has important historical value. Based on social network theory and methods, this article uncovers the correlation between word items in Dianshizhai Pictorial to reveal what the news pictorials were concerned about in Late Qing Dynasty. ［Method/process］ Firstly, the corpus was constructed by using General Review of Dianshizhai Pictorial as the corpus source. Secondly, the co-word network was established with Dianshizhai Pictorial inscriptions and postscripts as semantic units. Then, the keyword themes were established through community segmentation. Finally, the temporal egocentric network was constructed with the core keywords in the themes to examine the semantic coevolution of the keywords. ［Result/conclusion］ According to the research results, Dianshizhai Pictorial presents seven core discourse themes: “overseas news” “Shanghai” “strange and bizarre stories” “Chinese women” “Chinese celebrities” “Suzhou” and “illness and disability”.Taking the term “Chinese women” as an example, the chronological egocentric network shows that judicial terms such as “crime”, “murder case” appear more frequently and are evenly distributed, while “marriage” and “concubinage”, as indicators of old and new concepts, show the contrast in strength, which reflects the coexistence and transformation of old and new concepts in the hedge of Chinese and Western cultures. The results also show that social network method can effectively make up for some shortcomings of traditional research methods.
Hits
Hits
Downloads
Comment
YES