| 基于文本挖掘的人工智能科学主题演进研究 |
| |
| 引用本文： | 李牧南,王雯殊.基于文本挖掘的人工智能科学主题演进研究[J].情报杂志,2020,39(6):82-88. |
| |
| 作者姓名： | 李牧南 王雯殊 |
| |
| 作者单位： | 华南理工大学工商管理学院 广州 510641;华南理工大学工商管理学院 广州 510641 |
| |
| 基金项目： | 教育部哲学社会科学研究重大课题;国家自然科学基金;广东省科技计划;广东省软科学研究计划;中央高校基本科研业务费专项 |
| |
| 摘 要： | 目的/意义]人工智能相关的科学主题已经逐渐扩散到众多科学领域,这也导致人工智能科学的科学外延和学术边界不断被拓展,其分支主题也处于动态演进中;因此,分析人工智能科学主题演变就具有较为重要的情报和管理意义。方法/过程]为了进一步呈现和绘制人工智能相关科学主题的演进模式,一个基于LDA(Latent Dirichlet Allocation)和主题邻近度计算的文本挖掘方法被提出,尝试从文本建模视角呈现人工智能科学主题的演变趋势。结果/结论]通过采集超过22万篇与人工智能相关的研究文献,人工智能主题的演进模式得到了部分刻画和分析,这对于当前人工智能相关研究热点的预测与评估,以及相关政策制定或许具有一定的参考价值。
|
| 关 键 词： | 人工智能 主题建模 文本挖掘 科学主题演化 LDA |
Research on the Topic Evolution of Artificial Intelligence Based on Text Mining |
| |
| Li Munan,Wang Wenshu.Research on the Topic Evolution of Artificial Intelligence Based on Text Mining[J].Journal of Information,2020,39(6):82-88. |
| |
| Authors: | Li Munan Wang Wenshu |
| |
| Institution: | (School of Business Administration,South China University of Technology,Guangzhou 510641) |
| |
| Abstract: | Purpose/Significance]The scientific topics related to artificial intelligence have gradually spread to many academic fields,which has led to the continuous extension and expansion of the boundary of artificial intelligence,and the dynamic evolution of its sub-topics.Therefore,it is of great significance to analyze the evolution of the scientific topics of artificial intelligence.Method/Process]In order to further present and draw the evolution pattern of AI related scientific topics,a text mining method based on LDA(Latent Dirichlet Allocation)and topic-proximity calculation is proposed,trying to present the evolution trend of AI scientific topics from the perspective of text modeling.Result/Conclusion]Through the collection of more than 220,000 AI-related research literatures,the evolution patterns of AI topics have been partially described and analyzed,which may have certain enlightening significance and reference value for the prediction and evaluation of AI-related research hotspots,as well as the formulation of relevant science and technology policies. |
| |
| Keywords: | artificial intelligence topic modeling text mining scientific-topic evolution LDA |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
