[目的/意义] 在应用大数据提升图书馆服务的背景下,为提高借阅推荐的有效性,设计一种基于加权借阅网络的个性化推荐算法。[方法/过程] 引入复杂网络理论的分析方法,以深圳大学城图书馆的实际借阅数据为例,证明借阅网络具有二分图特性和BA无标度网络特性,具备个性化推荐的条件;通过构建读者借阅子网,采用能量传递六步法,实现个性化图书借阅推荐。[结果/结论] 实际借阅数据的验证结果表明,算法是有效且具有实际应用意义的。
[Purpose/significance] Under the background of improving library service with big data, in order to increase the effectiveness of borrowing recommendation, this paper designs a personalized recommendation algorithm based on weighted book-borrowing network.[Method/process] It introduces the analysis method of complex network theory. Then it takes the actual borrowing data of University Town of Shenzhen for example, proves that borrowing network has the characteristics of bipartite graph and BA scale-free network and has the condition of personalized recommendation. Finally, it takes energy transfer program to realize the personalized library recommendation.[Result/conclusion] The verification result of actual borrowing data shows that the algorithm is effective and has practical significance.
[1] REDNER S. How popular is your paper? An empirical study of the citation distribution[J]. The European physical journal b-condensed matter and complex systems,1998,4(2):131-134.
[2] ALBERT R,JEONG H, BARABUSI A L. Diameter of the World-Wide Web[J]. Nature,1999,401(6):130-131.
[3] JEONG H,TOMBOR B,ALBERT R,et al.The large-scale organization of metabolic networks[J]. Nature,2000,407(6804):651-654.
[4] JEONG H,MASON S,BARABUSI A L,et al. Lethality and centrality in protein networks[J]. Nature,2001,411(6833):41-42.
[5] 李楠楠, 张宁. 图书馆借阅网的二分图研究[J]. 复杂系统与复杂性科学, 2009,6(2):33-39.
[6] 马杰良, 宋艳, 潘贞贞,等. 图书漂流网络模型实证研究[J]. 计算机科学, 2015,42(3):51-54.
[7] 尚明生. 复杂网络及在情报工作中的应用[R/OL].[2015-11-19]. http://www.clas.ac.cn/xwzx/gzdt/201511/t20151105_4454419.html.
[8] CHARTRAND G, ZHANG P. Chromatic graph theory[M].Boca Paton:Taylor & FrancisInc, 2008.
[9] BARABASI A L, ALBERT R. Emergence of scaling in random networks[J].Science,1999,286(5439):509-512.
[10] 汪小帆,李翔, 陈关荣. 复杂网络理论及其应用[M].北京:清华大学出版社,2006.
[11] COHEN R,HAVLIN S. Scale-free networks are ultrasmall[J]. Physical review letters,2003,90(5):8701.
