论文部分内容阅读
【目的】研究中图法与DDC类目自动映射的问题,通过实现两者的互操作以达到集成检索、浏览和下载跨语言、跨地区的信息资源的目的。【方法】基于人工匹配映射的数据,研究基于特征集、类目匹配规则、类目关系、书目记录的语义匹配算法。【结果】实验证明,将近80%的类目与人工映射结果的数据相同,一定程度上提高了自动映射的准确性。【局限】仅基于人工映射经验提出基于特征词集的类目相似度计算,尚未实现语义层面的精确计算。实验数据局限在理学类目领域,未对其他领域的类目特征详细研究。【结论】综合考虑类目的含义受类名、类目注释、主题词、上下位关系等多种因素影响而提出的自动映射算法,相比当前主要考虑单一方面的要素而言,具有客观性和全面性。
【Objective】 The purpose of this paper is to study the automatic mapping between Chinese map method and DDC category, and to realize the goal of integrated retrieval, browsing and downloading of cross-regional and regional information resources through the interoperation between them. 【Method】 Based on the data of artificial matching mapping, the semantic matching algorithm based on feature set, category matching rule, category relationship and bibliographic record was studied. 【Result】 Experimental results show that nearly 80% of the categories have the same data as the manually mapped results, which improves the accuracy of automatic mapping to a certain extent. [Limitations] The category similarity calculation based on the feature word sets is based on the experience of manual mapping only, and the precise calculation at the semantic level has not been realized yet. The experimental data are limited to the area of Neo-Confucianism, and the characteristics of the other areas are not studied in detail. 【Conclusion】 The automatic mapping algorithm proposed by considering various factors such as class name, category annotation, keywords, upper and lower bounds, etc., has the characteristics of being objectivity Sexuality and comprehensiveness.