基于文本挖掘的图书采访推荐清单自动生成模型

摘要

由于大多数图书馆的采编馆员人数有限，依靠传统的手工统计表单方式采访图书往往很难令读者满意。运用Web的文本挖掘技术提取读者历史查询关键词，构造一种图书采访推荐清单的自动生成模型。实验数据证明，该模型有着较好的召回率和精准率，能有效地执行清单的自动生成。%For most libraries, there is only limited number of acquisitioning and cataloging librarians. Using traditional manual statistics form methods, the acquisitioning result is often far from satisfying the readers’ needs. Based on web text mining technology, by extracting history query keywords used by readers, the paper gives an automatic model of book-acquisitioning recommendation list. Experiments show that the model has better recall rates and accuracy rates, and the effective book-acquisitioning list can be automatically generated.