[1]夏 翔,刘 姜*,倪 枫,等.基于隐式反馈和加权用户偏好的推荐算法[J].计算机技术与发展,2024,34(03):140-146.[doi:10. 3969 / j. issn. 1673-629X. 2024. 03. 021]
　XIA Xiang,LIU Jiang*,NI Feng,et al.Recommendation Algorithms Based on Implicit Feedback and Weighted User Preferences[J].,2024,34(03):140-146.[doi:10. 3969 / j. issn. 1673-629X. 2024. 03. 021]
点击复制

基于隐式反馈和加权用户偏好的推荐算法()
分享到：

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

卷:
34
期数:
2024年03期
页码:
140-146
栏目:
人工智能
出版日期:
2024-03-10

文章信息/Info

Title:
Recommendation Algorithms Based on Implicit Feedback and Weighted User Preferences
文章编号:
1673-629X(2024)03-0140-07
作者:
夏 翔; 刘 姜* ; 倪 枫; 陆劲宇
上海理工大学 管理学院,上海 200093
Author(s):
XIA Xiang; LIU Jiang* ; NI Feng; LU Jin-yu
School of Management,University of Shanghai for Science and Technology,Shanghai 200093,China
关键词:
推荐算法; 隐式反馈; 操作频次; 用户偏好; 音乐推荐
Keywords:
recommendation algorithms; implicit feedback; frequency of operation; user preferences; music recommendation
分类号:
TP181
DOI:
10. 3969 / j. issn. 1673-629X. 2024. 03. 021
摘要:
针对现有隐式反馈算法中正负样本划分不合理、忽略用户操作频次、无法准确建模用户偏好等问题,提出一种基于隐式反馈和加权用户偏好的推荐算法( IFW-LFM) 。 该算法考虑了用户操作频次与正负样本划分间的关系,学习并改进wALS 算法,根据用户操作频次从缺失值中重新挖掘潜在正负样本,将用户操作频次大于 1 时的样本设置为正样本,用户操作频次为 1 或 0 时的样本为正样本或负样本,不再需要人为引入负样本;根据用户操作频次对用户偏好程度的影响,定义了置信度,明确用户偏好,并将其应用在隐因子模型的框架中;利用用户收听歌曲起止时间、收听时长等隐式反馈数据,提高隐式反馈样本利用度。 在两个音乐数据集上的对比实验结果说明,该算法在准确率、召回率与 NDCG 值上与 5 个经典隐式反馈算法( UserCF、ItemCF、LFM 、BPR、SVD) 相比最大平均提升了 45. 81% ,83. 83% 和 60. 33% ,具有更优的推荐效果。
Abstract:
In view of the unreasonable division of positive and negative samples,ignoring the frequency of user operations and failing toaccurately model user preferences,we propose a recommendation algorithm based on implicit feedback and weighted user preferences( IFW-LFM) . The algorithm considers the relationship between user operation frequency and positive and negative sample division,learns and improves wALS algorithm, and re - mines potential positive and negative samples from missing values according to user　operation frequency. It sets the samples with user operation frequency greater than 1 as positive samples and those with user operation frequency of 1 or 0 as positive or negative samples,eliminating the need to artificially introduce negative samples; defines the confidencelevel according to the influence of user operation frequency on the degree of user preference,specifies the user preference and applies it tothe framework of the hidden factor model;uses the user listening to song start and end time,listening duration and other implicit feedbackdata to improve the utilisation of implicit feedback samples. The results of the comparison experiments on two music datasets illustratethat the accuracy,　recall and NDCG values of the proposed method have a maximum average improvement of 45. 81% ,83. 83% and　60．33% respectively compared with the five classical implicit feedback algorithms of UserCF,ItemCF,LFM,BPR and SVD,which hasbetter recommendation results.

相似文献/References:

[1]李克潮,蓝冬梅.一种属性和评分的协同过滤混合推荐算法[J].计算机技术与发展,2013,(07):116.
　LI Ke-chao,LAN Dong-mei.A Collaborative Filtering Hybrid Recommendation Algorithm for Attribute and Rating[J].,2013,(03):116.
[2]王全民,王莉,曹建奇. 基于评论挖掘的改进的协同过滤推荐算法[J].计算机技术与发展,2015,25(10):24.
　WANG Quan-min,WANG Li,CAO Jian-qi. Improved Collaborative Filtering Recommendation Algorithm Based on Comments Mining[J].,2015,25(03):24.
[3]李远博,曹菡. 基于PCA降维的协同过滤推荐算法[J].计算机技术与发展,2016,26(02):26.
　LI Yuan-bo,CAO Han. Collaborative Filtering Recommendation Algorithm Based on PCA Dimension Reduction[J].,2016,26(03):26.
[4]张春生[],图雅[],翁慧[],等. 基于电子商务同类商品的推荐算法研究[J].计算机技术与发展,2016,26(05):17.
　ZHANG Chun-sheng[],TU Ya[],WENG Hui[],et al. Research on Similar Products Recommendation Algorithm Based on Electronic Commerce[J].,2016,26(03):17.
[5]房旋[],陈升波[],宫婧[][],等. 基于社交影响力的推荐算法[J].计算机技术与发展,2016,26(06):31.
　FANG Xuan[],CHEN Sheng-bo[],GONG Jing[][],et al. A Recommendation Algorithm Based on Social Influence[J].,2016,26(03):31.
[6]蒋宗礼,王威,陆晨. 基于均值预估的协同过滤推荐算法改进[J].计算机技术与发展,2017,27(05):1.
　JIANG Zong-li,WANG Wei,LU Chen. 基于均值预估的协同过滤推荐算法改进[J].,2017,27(03):1.
[7]许德心,李玲娟.基于 Spark 的关联规则挖掘算法并行化研究[J].计算机技术与发展,2019,29(03):30.[doi:10.3969/ j. issn.1673-629X.2019.03.006]
　XU De-xin,LI Ling-juan.Research on Parallelization of Association Rules Mining Algorithm Based on Spark[J].,2019,29(03):30.[doi:10.3969/ j. issn.1673-629X.2019.03.006]
[8]翟航天,汪学明.基于隐式反馈 LDA 模型的协同推荐算法研究[J].计算机技术与发展,2019,29(06):7.[doi:10. 3969 / j. issn. 1673-629X. 2019. 06. 002]
　ZHAI Hang-tian,WANG Xue-ming.Research on Collaborative Recommendation Algorithm Based on Implicit Feedback LDA Model[J].,2019,29(03):7.[doi:10. 3969 / j. issn. 1673-629X. 2019. 06. 002]
[9]史 伟,张青云,李 帅.基于 LBS 的停车位共享平台设计[J].计算机技术与发展,2019,29(11):149.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 030]
　SHI Wei,ZHANG Qing-yun,LI Shuai.Design of Sharing Platform for Parking Spaces Based on LBS[J].,2019,29(03):149.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 030]
[10]康嘉钰,苏凡军.基于生成对抗网络的长短兴趣推荐模型[J].计算机技术与发展,2020,30(06):35.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 007]
　KANG Jia-yu,SU Fan-jun.A Long-short-term Interests Recommendation Model Based on Generative Adversarial Networks[J].,2020,30(03):35.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 007]

更新日期/Last Update: 2024-03-10