论文部分内容阅读
【目的/意义】提出一种TF*IDF改进算法,用于全文分词后的语词权重计算,提取高权重语词,分析学科研究热点。【方法/过程】以万方数据库中2015年《情报学报》的载文为例,对每篇文章全文分词,用改进的TF*IDF方法计算语词权重。【结果/结论】发现该改进算法准确可行,且运用该方法分析得到,用户研究、大数据、情报学、社交网络、技术领域、文献作者、突发事件、零被引等,是2015年情报学的研究热点。
[Purpose / Significance] An improved TF * IDF algorithm is proposed for word weight calculation after full-text word segmentation, extracting high-weight words and analyzing hot topics in the field of research. [Method / Procedure] Taking the article published in the “Intelligence Journal” in 2015 in Wanfang Database for an example, the full-text participle of each article is used to calculate the word weight by using the improved TF * IDF method. [Results / Conclusion] The improved algorithm is found to be accurate and feasible. According to this method, user research, big data, information science, social networks, technical fields, literature authors, emergencies, Study hot spots.