一种K-Means改进算法在图书馆主题挖掘中的应用研究  

Research on the Application of K-Means Improved Algorithm in Library Topic Mining

作　　者：宋楚平[1] 李少芹[1] 

机构地区：[1]南通纺织职业技术学院,江苏南通226007

出　　处：《情报理论与实践》2014年第11期120-123,共4页Information Studies:Theory & Application

摘　　要：在现有图书借阅数据的基础上,对图书馆进行主题挖掘,来应对主动服务读者的要求。为减少主观因素对数据分析的影响,提高分析质量,采用传统K均值算法对图书馆主题挖掘是一种常用方法,但该算法本身存在一些固有的缺陷。为了改善图书馆主题挖掘效果,提出了一种基于K均值的改进算法。文章采用南通纺织职业技术学院1年的图书借阅数据对该算法和K均值算法进行了主题挖掘实验。结果表明,该算法在聚类准确度和收敛速度方面,相比K均值算法效果更好,聚类结果也更为合理。Based on the existing data of book-lending, library topic mining is carried out to meet the needs of the active service for readers. In order to reduce the influence of subjective factors on data analysis and improve the quality of analysis, it is a common method to use the traditional K-Means algorithm for library topic mining. However, the algorithm itself has some inherent defects. In order to improve the effect of library topic mining, an improved algorithm based on K-Means is proposed. This paper carries out the topic mining experiment for this improved algorithm and K-Means algorithm based on the 1 year＇ s book-lending data in our university. The results show that this improved algorithm in the aspects of clustering accuracy and convergence speed is better than K-Means algorithm and its clustering results are more reasonable.

关 键 词：K均值算法 主题挖掘 遗传算法 读者分类 

分 类 号：G254[文化科学—图书馆学]