[1]孙晶晶,荀亚玲,杨海峰.基于用户影响力和偏好一致性的社会化推荐[J].计算机技术与发展,2023,33(09):91-97.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 014]
 SUN Jing-jing,XUN Ya-ling,YANG Hai-feng.Social Recommendation Based on User Influence andPreference Consistency[J].,2023,33(09):91-97.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 014]
点击复制

基于用户影响力和偏好一致性的社会化推荐()
分享到:

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年09期
页码:
91-97
栏目:
人工智能
出版日期:
2023-09-10

文章信息/Info

Title:
Social Recommendation Based on User Influence andPreference Consistency
文章编号:
1673-629X(2023)09-0091-07
作者:
孙晶晶荀亚玲杨海峰
太原科技大学 计算机科学与技术学院,山西 太原 030024
Author(s):
SUN Jing-jingXUN Ya-lingYANG Hai-feng
School of Computer Science and Technology,Taiyuan University of Science and Technology,Taiyuan 030024,China
关键词:
社会化推荐综合信任协同过滤偏好一致性用户影响力
Keywords:
social recommendationcomprehensive trustcollaborative filteringpreference consistencyuser influence
分类号:
TP311
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 09. 014
摘要:
用户和项目的急剧增加使得评分数据过于稀疏导致传统推荐算法效果较差,社交网络信息的引入缓解了传统推荐系统中面临的数据稀疏性问题。 然而,现有社会化推荐在刻画用户之间的信任关系时未考虑到用户之间的信任具有偏好差异性和信任传播稳定性不强等问题。 因此,提出一种基于用户影响力和偏好一致性的社会化推荐。 首先,结合评分信息和社交信息从偏好一致性方向刻画用户之间的信任强度,挖掘出隐藏的信息,缓解了用户的偏好差异性。 其次,借助用户的社会影响力找到一条信任传播稳定性最强的路径,避免信任在传播过程中造成信任节点信息的丢失。 然后,将用户的评分相似度和信任相似度线性加权得到用户的近邻用户做评分预测。 最后,将该方法与现有社会化推荐算法在Filmtrust 和 CiaoDVD 数据集上进行综合实验,结果表明该方法在 MAE 和 RMSE 上优于现有推荐算法。
Abstract:
The sharp increase of users and items makes the rating data too sparse,which leads to poor performance of traditional recommendation algorithms. The introduction?
of social network information alleviates the data sparsity problem faced by traditionalrecommendation systems. However,the existing social recommendation does not take into account the differences in preferences betweenusers and the weak stability of trust propagation when describing the trust relationship between users. Therefore,a social recommendationbased on user influence and preference consistency is proposed. Firstly,combining the rating information and social information,the truststrength between users is described from the direction of preference consistency,and the hidden information is mined to alleviate the user’spreference difference. Secondly,with the social influence of users,a path with the strongest trust propagation stability is found,whichavoids the loss of trust node information in the process of trust propagation. Then,the user’s rating similarity and trust similarity arelinearly weighted to obtain the user’ s neighbors for rating prediction. Finally, the proposed method and the existing socialrecommendation algorithm are comprehensively tested on the Filmtrust and CiaoDVD datasets. It is showed that the proposed method outperforms the existing recommendation algorithms in MAE and RMSE.
更新日期/Last Update: 2023-09-10