论文部分内容阅读
[目的 /意义]基于科技论文多特征项共现突发强度的分析方法研究是将各学科领域科技论文文献载体中的多特征项共现信息定量化、重点热点突发的信息内容可视化的知识图谱分析方法。从动态论文等文献的文档流中探测出突发的特征项对识别密集的内容、活跃的特征项以及预测文本内容的发展走势具有重要的意义。[方法 /过程]本研究针对科技论文多特征项共现的突发监测问题,对比目前已有的突发监测分析算法,将改进后的基于卡方统计的热点词计算方法进一步应用于本研究所设计的多特征项突发共现分析方法,并自主开发多特征项突发共现可视化分析工具,用于科技论文多特征项突发共现的图谱可视化,以期通过该研究来揭示相关科技文献的变化状况及突发的热点内容。[结果 /结论]通过将本方法应用到科研机构年度发表论文的监测当中,可以监测分析科研机构发文作者、关键词、发表期刊及其相互间关系(如作者-关键词等)在各年的突发情况,并能通过该分析方法以及交叉图谱进一步解读突发特征项的含义,并能揭示出比分析单一特征项突发情况更为广泛和深入的知识内容。
[Purpose / Significance] Based on the analysis of the co-occurrence burst strength of multi-feature items in scientific papers, the research is to quantify the co-occurrence information of multi-feature items in the scientific and technical paper document carriers of various disciplines and focus on the hot and sudden knowledge of information contents visualization Graph analysis method. The detection of sudden characteristic items from the document flow of documents such as dynamic essays is of great significance for the identification of intensive content, active characteristic items and the trend of predicting text content. [Method / Process] This study aims at the problem of sudden monitoring of co-occurrence of multi-feature items in scientific papers. Comparing the existing algorithms of sudden monitoring and analysis, we apply the improved method based on chi-square statistics to calculate the hot words in this study The multi-feature item is designed to burst co-occurrence analysis method, and the multi-feature item burst co-occurrence visualization analysis tool is developed independently to visualize the burst co-occurrence graph of multi-feature items in the paper so as to reveal the related technology Literature changes and sudden hot topics. [Results / Conclusion] By applying this method to the annual papers published by scientific research institutes, it is possible to monitor and analyze the authors, keywords, published journals and their relationships (such as author-key words) of scientific research institutes in each year And can further interpret the meaning of the sudden characteristic items through the analysis method and the cross-map, and can reveal more extensive and deeper knowledge contents than the analysis of the sudden onset of the single characteristic term.