基于随机森林的图书采购决策模型  

The Model of Book Purchasing Decision-making Based on Random Forest

作　　者：傅立云[1] 胡芸 Fu Liyun;Hu Yun(Xiangtan University Library,Xiangtan Hunan 411105;School of Management,Xiangtan University,Xiangtan Hunan 411105)

机构地区：[1]湘潭大学图书馆,湖南湘潭411105 [2]湘潭大学管理学院,湖南湘潭411105

出　　处：《情报探索》2020年第5期34-39,共6页Information Research

基　　金：湖南省高校教改项目“基于SPOC模式的研究生文献检索课程建设研究”(项目编号:湘教通〔2018〕164号)成果之一。

摘　　要：[目的/意义]为了最大限度地满足读者需求和优化馆藏的需要,辅助馆员进行图书采购决策。[方法/过程]综合考量图书采购的多种影响因素,对图书馆藏书和借阅数据进行挖掘,借助人工智能技术的研究成果,运用随机森林构建了图书采购决策模型。以湘潭大学图书馆近10年的采购数据作为样本,进行实验对比分析,并对模型进行反复调整与改进,形成新的采购模型。[结果/结论]实验结果表明该模型建设思路切实可行,具有较好的准确性,能够有效提高图书采购的资金利用率,为图书馆进行采购决策提供新思路和途径。[Purpose/significance]The paper is to maximally meet the needs of readers and optimize the collection, and assist librarians to make book purchasing decisions.[Method/process] The paper comprehensively considers the various influencing factors of book purchasing to mine the library collection and borrowing data. With reference to the research results in AI, a book purchasing decision-making model is constructed by random forest. Then, taking the procurement data of Xiangtan University Library for nearly 10 years as a sample, the experiment is compared and analyzed, and the model is repeatedly adjusted and improved. [Result/conclusion]The experimental results show that the model constructed is feasible and accurate, which can effectively improve the utilization rate of funds and provide new ideas and approaches for library purchasing decision-making.

关 键 词：图书采购 随机森林 决策树 数字图书馆 

分 类 号：G251.5[文化科学—图书馆学]