文献老化在图书馆书目挖掘中的应用研究  

Application Research of Literature Obsolescence in the Bibliography Mining

作　　者：邓奇强[1] 熊燕[2] 郭锋[3] 

机构地区：[1]怀化学院图书馆技术部 [2]怀化学院图书馆文献资源建设部 [3]怀化学院图书馆

出　　处：《图书情报工作》2013年第6期91-96,共6页Library and Information Service

基　　金：湖南省教育厅科研项目"基于知识短缺的图书馆读者聚类定向挖掘研究"(项目编号:12C0847);怀化学院一般项目"基于关联规则的书目推荐服务研究"(项目编号:HHUY2010-58)研究成果之一

摘　　要：对怀化学院图书馆TP类图书按编目日期统计各年借阅量,依据文献老化定律、最小二乘法曲线拟合求得TP类图书的半衰期及老化曲线,利用TP类图书的老化曲线方程预测其借阅量,提出图书关联规则支持度的老化常数。实验表明:通过关联规则中的support(X)=support(X)×K’方法降低老化图书的关联支持度,减少老化图书相对较弱的强关联规则,解决了原书目推荐系统推荐给读者的书目中有相当一部分图书已失去使用价值的问题。此方法不仅适用于怀化学院图书馆馆藏图书的书目挖掘,对其他高校图书馆也有借鉴价值。By counting the amount of borrowing TP books in our library each year based on the cataloging date, this paper curves to determine half - life decay and aging curve of the TP books with the literature aging law and the least squares method. Then it predicts the amount of borrowing TP books by aging curve equation. This paper proposes the aging constant of the association rules support K＇ = Mtx / Mm （ Mtx is the amount of borrowing books of new cataloging in one year, Mm is the amount of new cataloging of books borrowing peak in a year）. The experiments show that it can reduce the associated support by the method of support （X） = support （X） × K＇ in association rules and the amount of relatively weak strong association rules of aging book, and solve the problem that a considerable part of the book original bibliographic recommendation system recommends to readers has lost its use value. The analysis shows that this method is not only suitable for the bibliographic mining of the collections, but makes the reference for the university libraries.

关 键 词：文献老化 图书老化 书目挖掘 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] G250.7[自动化与计算机技术—计算机科学与技术]