融合时间衰减与偏好波动的协同偏好获取方法

doi:10.11772/j.issn.1001-9081.2016.07.2011

计算机应用 ›› 2016, Vol. 36 ›› Issue (7): 2011-2015.DOI: 10.11772/j.issn.1001-9081.2016.07.2011

融合时间衰减与偏好波动的协同偏好获取方法

杨立¹, 胡运红², 邵桂荣¹

1. 运城学院公共计算机教学部, 山西运城 044000;
2. 运城学院应用数学系, 山西运城 044000

收稿日期:2016-01-08 修回日期:2016-03-02 出版日期:2016-07-10 发布日期:2016-07-14
通讯作者: 杨立
作者简介:杨立(1978-),男,山西运城人,讲师,硕士,主要研究方向:图像处理、人工智能;胡运红(1974-),女,山西运城人,副教授,博士,主要研究方向:最优化理论与算法、数据挖掘、机器学习;邵桂荣(1979-),女,山西省运城人,副教授,硕士,主要研究方向:智能控制算法。
基金资助:
国家自然科学基金资助项目（11241005）；山西省教育厅重点教学研究改革项目（J2014104）。

Preference prediction method based on time attenuation and preference fluctuation

YANG Li¹, HU Yunhong², SHAO Guirong¹

1. Public Department of Computer Teaching, Yuncheng University, Yuncheng Shanxi 044000, China;
2. Department of Applied Mathematics, Yuncheng University, Yuncheng Shanxi 044000, China

Received:2016-01-08 Revised:2016-03-02 Online:2016-07-10 Published:2016-07-14
Supported by:
This work is partially supported by the National Natural Science Foundation of China (11241005), Shanxi Provincial Education Department's Key Teaching and Research Reform Project (J2014104).

摘要/Abstract

摘要： 针对现有的推荐系统多采用近邻用户的偏好行为来预测当前用户的偏好，而不考虑用户的偏好会随着时间的变化而改变，影响了推荐准确率的问题，提出了一种基于时间衰减与偏好波动的协同偏好获取方法。首先，基于时间因素、用户历史偏好等获取偏好衰减增量与衰减速度，并据此生成衰减函数，使用衰减函数对用户历史行为数据进行衰减修正；其次，基于用户的历史偏好分布获取其偏好波动幅度；最后，将衰减函数与偏好波动幅度分别加入到最近邻获取与偏好获取流程，协同为用户生成推荐列表。在大规模真实数据集上的实验结果表明，所提出的方法与基于属性评分分布的协同过滤（RDCF）与最优Top-N的协同过滤（OTCF）相比，平均绝对误差（MAE）值分别降低了近6.42%和7.73%。实验结果表明所提方法能够提高推荐准确度，提升推荐质量。

关键词: 推荐系统, 时间衰减, 衰减增量, 衰减速度, 偏好波动

Abstract: The existing recommender systems often use the nearest neighbors' preference behavior to predict current users' preference, and their recommendation accuracy are influenced by the lack of consideration that users' preference would change over time. To solve this problem, a cooperative preference prediction method based on time attenuation and preference fluctuation was proposed. First, attenuation increment and attenuation speed were obtained based on time and historical preference, and the attenuation function was generated by attenuation increment and attenuation speed to modify users' historical preference behavior. Then the distribution of historical preference was used to compute the preference fluctuation range. Finally, the recommender list was generated for user by applying the attenuation function and preference fluctuation range into the acquisition of nearest neighbors and the preference acquisition process. The experimental results on real data set show that, compared with the Collaborative Filtering based on Rating Distribution (RDCF) and Optimizing Top-N Collaborative Filtering (OTCF), the average Mean Absolute Error (MAE) of the proposed method is decreased by about 6.42% and 7.73% respectively. It also shows that the proposed method can achieve higher recommendation accuracy and better recommendation quality.

Key words: recommender system, time attenuation, attenuation increment, attenuation speed, preference fluctuation

中图分类号:

TP391.9
TP18

杨立, 胡运红, 邵桂荣. 融合时间衰减与偏好波动的协同偏好获取方法[J]. 计算机应用, 2016, 36(7): 2011-2015.

YANG Li, HU Yunhong, SHAO Guirong. Preference prediction method based on time attenuation and preference fluctuation[J]. Journal of Computer Applications, 2016, 36(7): 2011-2015.

参考文献

[1] 吴月萍,杜奕.基于人工鱼群算法的协同过滤推荐算法[J].计算机工程与设计,2012,33(5):1852-1856.(WU Y P, DU Y. Collaboration filtering recommendation algorithm based on artificial fish swarm algorithm[J]. Computer Engineering & Design, 2012, 33(5):1852-1856.)
[2] PARK S T, CHU W. Pairwise preference regression for cold-start recommendation[C]//Proceedings of the Third ACM Conference on Recommender Systems. New York, ACM, 2009:21-28.
[3] DAI Y, DANIELS A, WU J. Demo:recommendation system for dynamic spectrum access through spectrum mining and prediction[J]. Spectrum, 2014, 43(5):415-416.
[4] CHEN W, HSU W, LEE M L. Modeling user's receptiveness over time for recommendation[C]//Proceeding of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, ACM, 2013:373-382.
[5] 汪静,印鉴,郑利荣,等.基于共同评分和相似性权重的协同过滤推荐算法[J].计算机科学,2010,37(2):99-104.(WANG J, YIN J, ZHENG L R, et al. Collaborative filtering recommendation algorithm based on co-ratings and similarity weight[J]. Computer Science, 2010, 37(2):99-104.)
[6] 郭晶晶,马建峰.面向虚拟社区物联网的信任推荐算法[J].西安电子科技大学学报(自然科学版),2015,42(2):59-65.(GUO J J, MA J F. Trust recommendation algorithm for virtual community based interest of things[J]. Journal of Xidian University (Natural Science), 2015, 42(2):59-65.)
[7] 孙光福,吴乐,刘淇,等.基于时序行为的协同过滤推荐算法[J].软件学报,2013,24(11):2711-2733.(SUN G F, WU L, LIU Q, et al. Recommendations based on collaborative filtering by exploiting sequential behaviors[J]. Journal of Software, 2013, 24(11):2711-2733.)
[8] 王兴茂,张兴明.基于贡献因子的协同过滤推荐算法[J].计算机应用研究,2015,32(12):132-136.(WANG X M, ZHANG X M. Collaborative recommendation algorithm based on contributor factor[J]. Application Research of Computers, 2015, 32(12):132-136.)
[9] FOUSS F, PIROTTE A, RENDERS M, et al. Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation[J]. Knowledge and Data Engineering, 2007, 19(3):355-369.
[10] 李源鑫,肖如良,陈洪涛,等.时间衰减制导的协同过滤相似性计算[J].计算机系统应用,2013,22(11):129-134.(LI Y X, XIAO R L, CHEN H T. Time decay guided similarity calculation in collaborative filtering[J]. Computer Systems & Applications, 2013, 22(11):129-134.)
[11] 王茜,杨莉云,杨德礼.面向用户偏好的属性值评分分布协同过滤算法[J].系统工程学报,2010,25(4):561-568.(WANG Q, YANG L Y, YANG D L. Collaborative filtering algorithm based on rating distribution of attributions faced user preference[J]. Journal of Systems Engineering, 2010, 25(4):561-568.)
[12] ZHANG W, CHEN T, WANG J, et al. Optimizing top-N collaborative filtering via dynamic negative item sampling[C]//Proceeding of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, ACM, 2013:785-788.

[1]	刘欢, 李晓戈, 胡立坤, 胡飞雄, 王鹏华. 基于知识图谱驱动的图神经网络推荐模型[J]. 计算机应用, 2021, 41(7): 1865-1870.
[2]	林怿星, 唐华. 基于异构信息网络的混合推荐模型[J]. 计算机应用, 2021, 41(5): 1348-1355.
[3]	高铭蔚, 桑楠, 杨茂林. 基于胶囊网络的交互式网络电视视频点播推荐模型[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3171-3177.
[4]	任柯舟, 彭甫镕, 郭鑫, 王喆, 张晓静. 动态融合社交信息的社会化推荐[J]. 计算机应用, 2021, 41(10): 2806-2812.
[5]	李翔锟, 贾彩燕. 融合重叠社区正则化及隐式反馈的协同过滤方法[J]. 计算机应用, 2021, 41(1): 53-59.
[6]	李春英, 汤庸, 肖政宏, 李天送. 学术社交网络中的权威学者推荐模型[J]. 计算机应用, 2020, 40(9): 2594-2599.
[7]	秦婧, 张青博, 王斌. 关注长尾物品的推荐方法[J]. 计算机应用, 2020, 40(2): 454-458.
[8]	张文龙, 钱付兰, 陈洁, 赵姝, 张燕平. 基于双重最相关注意力网络的协同过滤推荐算法[J]. 计算机应用, 2020, 40(12): 3445-3450.
[9]	雷曼, 龚琴, 王纪超, 王保群. 基于标签权重的协同过滤推荐算法[J]. 计算机应用, 2019, 39(3): 634-638.
[10]	田保军, 杨浒昀, 房建东. 融合信任和基于概率矩阵分解的推荐算法[J]. 计算机应用, 2019, 39(10): 2834-2840.
[11]	文凯, 朱传亮. 融合社交网络和兴趣的正则化矩阵分解推荐模型[J]. 计算机应用, 2018, 38(9): 2523-2528.
[12]	杨帅, 王鹃. 基于堆栈降噪自编码器改进的混合推荐算法[J]. 计算机应用, 2018, 38(7): 1866-1871.
[13]	李全, 刘兴红, 许新华, 林松. 基于联合概率矩阵分解的个性化试题推荐方法[J]. 计算机应用, 2018, 38(3): 639-643.
[14]	杨丰瑞, 郑云俊, 张昌. 结合概率矩阵分解的混合型推荐算法[J]. 计算机应用, 2018, 38(3): 644-649.
[15]	孙丽梅, 李悦, Ejike Ifeanyi Michael, 曹科研. 简化的Slope One在线评分预测算法[J]. 计算机应用, 2018, 38(2): 497-502.

融合时间衰减与偏好波动的协同偏好获取方法

Preference prediction method based on time attenuation and preference fluctuation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics