当前位置>主页 > 期刊在线 > 信息技术 >

信息技术2020年1期

基于谱聚类和LFM 的选课推荐算法设计
刘旋
(信阳农林学院 信息工程学院,河南 信阳 464000)

摘  要:高校教务系统中学生数量和课程种类的飞速增长,使得传统推荐算法难以处理海量、高维的选课数据,为进一步提升大学生的选课效率,文章提出一种改进的LFM 隐语义模型推荐算法,首先构造选课评分数据的相似矩阵,通过谱聚类进行初始分类,然后分类别构建LFM 模型并计算合理的推荐算法。通过在某高校的选课数据集上的对比实验,证明了本文算法具有较高的预测精度和较低的空间复杂度。


关键词:推荐算法;隐语义模型;谱聚类算法



中图分类号:TP391         文献标识码:A       文章编号:2096-4706(2020)01-0014-03


A Recommended Courses Algorithm Based on Spectral Clustering and LFM

LIU Xuan

(College of Information Engineering,Xinyang Agriculture and Forestry University,Xinyang 464000,China)

Abstract:The rapid growth of the number of students and the types of courses in the educational administration system of colleges and universities,make the traditional recommendation algorithm is difficult to deal with mass and course of high-dimensional data,in order to further enhance students’course selection efficiency,this paper proposes a recommendation algorithm to improve the LFM argot meaning of model,the first data structure course score of similar matrix,the initial classification by spectral clustering and classification build LFM model and calculate the reasonable recommendations. Through the comparison experiment on the data set of course selection in a university,it is proved that the algorithm in this paper has higher prediction accuracy and lower space complexity.

Keywords:recommendation algorithm;LFM(latent factor model);spectral clustering algorithm


基金项目:河南省教育厅人文社会科学研究一般项目(2020-ZDJH-353);信阳市哲学社会科学项目(2019JY048)


参考文献:

[1] 沈苗,来天平,王素美,等. 北京大学课程推荐引擎的设计和实现 [J]. 智能系统学报,2015,10(3):369-375.

[2] 陈万志,张爽,王德建,等. 基于近邻模型与概率矩阵分解的高校选课推荐算法 [J]. 辽宁工程技术大学学报(自然科学版),2017,36(9):976-982.

[3] GANU G,KAKODKAR Y,MARIAN A. Improving the quality of predictions using textual information in online user reviews [J].Information Systems,2013,38(1):1-15.

[4] KOOPMAN S J,LUCAS A,MONTEIRO A A. The Multi-State Latent Factor Intensity Model for Credit Rating Transitions [J].SSRN Electronic Journal,2008,142(1):399-424.

[5] 陈晔,刘志强. 基于LFM 矩阵分解的推荐算法优化研究 [J]. 计算机工程与应用,2019,55(2):116-120+167.

[6] DHILLON I S,GUAN Y,KULIS B. Kernel k-means:spectral clustering and normalized cuts [C]//KDD’04 Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining,Seattle,USA,August 22-25,2004. New York:ACM,2004:551-556.

[7] WANG J,VRIES A P D,REINDERS M J T. Unifying userbased and item-based collaborative filtering approaches by similarity fusion [C]//SIGIR 2006:Proceedings of the 29th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval,Seattle,Washington,USA,August 6-11,2006. ACM,2006.


作者简介:刘旋(1991-),男,汉族,河南信阳人,讲师,硕士,研究方向:数据挖掘、模式分析。