| 学科主题演化路径的多模式识别与预测——一个情报学学科主题演化案例 |
| |
| 引用本文： | 隗玲,许海云,胡正银,董坤,王超,庞弘燊.学科主题演化路径的多模式识别与预测——一个情报学学科主题演化案例[J].图书情报工作,2016,60(13):71-81. |
| |
| 作者姓名： | 隗玲 许海云 胡正银 董坤 王超 庞弘燊 |
| |
| 作者单位： | 1. 山西财经大学信息管理学院 太原 030006;
2. 中国科学院成都文献情报中心 成都 610041;
3. 中国科学院大学 北京 100049;
4. 中国科学院广州生物医药与健康研究院 广州 510530 |
| |
| 基金项目： | 本文系国家自然科学基金管理学部青年项目“基于学科领域科技论文多重共现的情报计量分析方法研究”（项目编号：71403263）研究成果之一。 |
| |
| 摘 要： | 目的/意义] 基于主题关联相似度揭示主题汇聚及变异过程，识别学科交叉主题及交叉模式，归纳学科主题的演化趋势及演化路径模式。方法/过程] 获取情报学学科科研论文的高频主题词，构造主题词共词矩阵，利用网络社区演化分析工具生成学科主题演化网络图，结合指标数据对学科主题演化过程进行分析。结果/结论] 总体上看，情报学学科的研究主题虽然在反复地变化，但核心主题一直存在；扩张、收缩和合并是研究主题最普遍的变化态势，分裂现象较少，产生和消亡现象存在；有3条特定社区演化轨迹清晰地贯穿始终，活跃度相对稳定，反映了3类核心研究主题；3类核心研究主题的演化路径呈现出升华吸纳、共融迭新和辐射推进3种演化模式。研究结果显示，基于主题关联学科主题演化路径的多模式识别方法既能从宏观层面呈现学科主题演化形式，也能从微观层面分析学科主题交叉模式，结合二者可揭示学科主题的继承或创新，预测学科交叉主题的发展方向。
|
| 关 键 词： | 主题关联 主题演化 学科交叉 演化路径 |
| 收稿时间： | 2016-04-28 |
Multiple-pattern Analysis and Prediction of Topic Evolution Path Based on Topic Correlation: A Case Study of Information Science Research |
| |
| Wei Ling,Xu Haiyun,Hu Zhengyin,Dong Kun,Wang Chao,Pang Hongsen.Multiple-pattern Analysis and Prediction of Topic Evolution Path Based on Topic Correlation: A Case Study of Information Science Research[J].Library and Information Service,2016,60(13):71-81. |
| |
| Authors: | Wei Ling Xu Haiyun Hu Zhengyin Dong Kun Wang Chao Pang Hongsen |
| |
| Institution: | 1. School of Information and Management, Shanxi University of Finance and Economics, Taiyuan 030006;
2. Chengdu Documentation and Information Center, Chinese Academy of Sciences, Chengdu 610041;
3. University of Chinese Academy of Sciences, Beijing 100049;
4. Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530 |
| |
| Abstract: | Purpose/significance] Based on the topic correlation, this paper made a detailed analysis on the phenomenon and process of topic convergence or transformation, detected the interdisciplinary topics and their cross modality, and generalized the topics' evolution trend and multiple patterns of evolution path.Method/process] Firstly, high frequency keywords of information science papers were selected to generate a co-word matrix, which was transferred to be loaded into a community detection tool. Then, community evolution graphs and statistics were both used to analyze the topic evolution.Result/conclusion] On the whole, the topics changed constantly over time, but the core topics remain unchanged. The phenomenon of expansion, contraction and convergence are common; division is rare; emergence and disappearance are moderate. Three specific community evolution traces with relatively stable activity developed clearly throughout the time range, reflecting the three types of core topics' continuity. The evolution paths of the three kinds of core topics present three evolution models, that is sublimation-absorbing, inclusion-iteration, and radiation-promotion. According to the research, the method of multiple-pattern analysis and prediction of topic evolution based on topic correlation, can reveal the evolution patterns of topics in a macro level, and analyze the cross modalities of topics in a micro level. Combining the advantages of both, the method can uncover the inheritance or innovation of topics, and predict the future directions of interdisciplinary topics. |
| |
| Keywords: | topic correlation topic evolution interdisciplinary evolution path |
|
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
|
