期刊文献+

基于改进Slope One的大数据智能化推荐研究

Research on Intelligent Recommendation of Big Data Based on Improved Slope One
下载PDF
导出
摘要 传统推荐方法忽略了对大数据项目相似度的预测,导致推荐偏差较大。为此以Slope One为基础算法,提出新的大数据智能化推荐方法。基于使用者评分的Slope One算法,通过赋予各项目权值,优化Slope One算法。分别构建信任机制与皮尔森相关系数的Slope One优化算法,利用蚁群聚类算法,划分使用者属性特征类别。结合项目使用者属性相似度与使用者相似度的预测评分,得到使用者综合相似度,构建预测评分模型,对比预设评分阈值后,完成大数据智能化推荐。针对Epinions与Film Trust两个数据集,从均方根误差等角度,量化评估方法性能。实验结果表明,所提方法的推荐误差较小,具有较为理想的推荐效果。 In traditional recommendation methods, the lack of similarity prediction of big data items leads to large deviation. Therefore, this paper presents a big data intelligent recommendation method based on Slope One basic algorithm. In order to optimize the Slope One algorithm, the weight of each item was given according to the Slope One algorithm. The Slope One optimization algorithm for trust mechanism and Pearson correlation coefficient were constructed, respectively. Ant colony clustering algorithm was used to classify users’ attribute feature categories. The prediction score of user attribute similarity and user similarity was utilized to obtain the comprehensive similarity of users, constructing the prediction score model. The preset score was compared to complete the intelligent recommendation of big data. The performance of the evaluation method was quantified according to the root mean square error for the two data sets of Espions and Film Trust. The experimental results show that this method has small recommendation error and good recommendation effect.
作者 徐永华 雷东升 XU Yong-hua;LEI Dong-sheng(School of Computer Engineering,Jinling University of Science and Technology,Nanjing Jiangsu 211169,China;Beijing University of Technology,Beijing 100124,China)
出处 《计算机仿真》 北大核心 2022年第1期460-464,共5页 Computer Simulation
基金 教育部高等教育司2019年第一批产学合作协同育人项目(201901051055) 江苏省现代教育技术研究“十三·五”规划2018年度立项课题(2018-R-62636)。
关键词 信任机制 皮尔森相关系数 属性特征 相似度 Weighted Slope One Algorithm Trust Mechanism Pearson Correlation Coefficient Attribute Features Similarity
  • 相关文献

参考文献10

二级参考文献59

共引文献112

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部