基于JAVA的图书馆中文查新智能去重系统的研究与实现

Research and Realization of the an Intelligent Duplication Deleting System Based on Java

哈尔滨工业大学图书馆应用Java语言开发的《图书馆中文查新智能去重系统》,采用基于字段的主题全匹配方法,针对不同数据中采集数据样本可能存在的细微差异,利用句子相似度方法进行衡量,对高相似度结果进行二次校验整理,得到最终去重结果。查新员只需将各个数据库的导出结果导入系统就可轻松实现相同文献的去重,并可按照不同查新站的报告模板导出符合要求的文献格式,从而最大限度地节省查新员处理文献的时间,使得其将有限的查新时间更好地用于文献的对比分析,从而更好地提高查新质量。

Java language is used to develop an intelligent duplication deleting system for chinese novelty search in Harbin Institute of Technology Library. The system uses the full-matching method based on the theme of the field, collects different data sample, measures similarity of the same fields and second checking to achieve the final duplication deleting results. Novelty consultants only need to export the results of each database and easily achieve the function of deleting copies. Then according to the different report formats it can export automatically to meet the requirements of format. Consequently saving much time of dealing with literature, novelty consultants will have more time to compare and analyze literature in order to improve the quality of investigation.