A hybrid collaborative filtering recommendation model based on complex attribute of goods
-
摘要: 协同过滤作为应用最广、研究最多的推荐算法，但依旧面临数据稀疏性、冷启动、数据质量差等固有问题，同时也鲜有研究者从实用角度基于商品性价比方面提高预测精确度.为此，本文综合考虑用户主观评分和商品客观评分，并在此基础上结合情境预过滤、社会网络理论以及专家意见提出了一种混合协同过滤推荐模型，在一定程度上缓解了上述缺点.并通过真实网上汽车销售数据实验，表明该模型相对传统协同过滤具有更高的预测精度，更适用于具有复杂属性的商品.Abstract: Collaborative filtering as the most widely used, the most recommendation algorithm, the shortcomings inherent in the data sparse, cold startpoor data quality and others, and few studies based on commodity price to improve the prediction accuracy. At the same time, facing the full e-commerce market network Navy, the ratings and reviews also indirectly led to the predict a decline in accuracy. Therefore, this paper comprehensive consideration of the user subjective ratings and objective product score, and on this basis, combined with situation pre filtering, social network theory and expert opinions put forward a hybrid collaborative filtering recommendation model, to some extent alleviate the above shortcomings. And through experiment with real online car sales data, the model has higher forecast accuracy than the traditional collaborative filtering, and is more suitable for the commodity with complex attributes.
-
Key words:
- collaborative filtering /
- context /
- complex attribute /
- personalized recommendation
-
表 1 用户、商品矩阵
Tab. 1 User-item matrix
$ui_1$ $ui_2$ $ui_3$ $ui_4$ $ui_5$ $ui_6$ $ui_7$ $ui_8$ $ci_1$ $ci_2$ $ci_3$ $ci_4$ $ci_5$ $ci_6$ $ci_7$ $ci_8$ $u_1$ 1 5 3 3 4 5 5 4 1 5 3 3 4 5 5 4 $u_2$ 1 4 2 1 3 4 4 3 1 3 2 1 4 4 3 3 $u_3$ 2 5 3 3 4 5 5 4 1 4 2 2 2 3 3 1 $u_4$ 4 3 5 4 2 4 3 1 2 3 4 3 1 2 2 3 $u_5$ 5 5 5 5 5 5 5 5 1 4 5 4 5 1 3 4 
下载: 导出CSV
表 2 用户、偏好客观评分排序矩阵
Tab. 2 User, preference objective scoring matrix
$i_2 $ $i_6 $ $i_7 $ $i_5 $ $i_8 $ $i_3 $ $i_4 $ $i_1 $ $u_1 $ 5 5 5 4 4 3 3 1 $u_2 $ 4 3.6 3.6 3.4 3 2 1 1 $u_3 $ 4.6 4.2 4.2 3.2 2.8 2.6 2.6 1.6
下载: 导出CSV
表 3 用户偏好关系矩阵
Tab. 3 User, preference objective scoring presort matrix
$i_2 $ $i_6 $ $i_7 $ $i_5 $ $i_8 $ $i_3 $ $i_4 $ $i_1 $ $u_1 $ 1 1 1 1 1 0 0 0 $u_2 $ 1 1 1 1 1 0 0 0 $u_3 $ 1 1 1 0 0 0 0 0
下载: 导出CSV
表 4 用户偏好关系矩阵
Tab. 4 User, preferencerelation matrix
$u_1 $ $u_2 $ $u_3 $ $u_1 $ 0 5 3 $u_2 $ 5 0 3 $u_3 $ 3 3 0
下载: 导出CSV
表 5 标准化用户偏好关系矩阵
Tab. 5 Standared user, preference relation matrix
$u_1 $ $u_2 $ $u_3 $ $u_1 $ 0 5/8 3/8 $u_2$ 5/8 0 3/8 $u_3 $ 3/6 3/6 0
下载: 导出CSV
-
[1] 许海玲, 吴潇, 李晓东.互联网推荐系统比较研究[J].软件学报, 2009, 20(2):350-362. http://www.cnki.com.cn/Article/CJFDTOTAL-RJXB200902015.htm [2] 张锋, 常会友.使用BP神经网络缓解协同过滤推荐算法的稀疏性问题[J].计算机研究与发展, 2006, 43(4):667-672. http://www.cnki.com.cn/Article/CJFDTOTAL-JFYZ200604014.htm [3] 王辉, 高利军, 王听忠.个性化服务中基于用户聚类的协同过滤推荐[J].计算机应用, 2007, 27(5):1225-1227. http://www.cnki.com.cn/Article/CJFDTOTAL-JSJK200904034.htm [4] ZIEGLER C N, MCNEE S, KONSTAN J, et al. Improving recommendation lists through topic diversification[C]//Proceedings of the 14th International World Wide Web Conference. 2005:22-32. [5] 陈曦, 成韵姿.一种优化组合相似度的协同过滤推荐算法[J].计算机工程与科学, 2017, 39(1):180-187. http://www.cnki.com.cn/Article/CJFDTOTAL-JSJK201701025.htm [6] 郭彩云, 王会进.改进的基于标签的协同过滤算法[J].计算机工程与应用, 2016, 52(8):56-61. http://www.cnki.com.cn/Article/CJFDTOTAL-JSGG201608012.htm [7] 宋伟伟, 杨德刚, 郑敏.基于时间加权标签的协同过滤推荐算法研究[J].重庆师范大学学报(自然科学版), 2016, 33(5):113-120. http://www.cnki.com.cn/Article/CJFDTOTAL-CQSF201605023.htm [8] 孙楠军, 刘天时.基于项目分类和用户群体兴趣的协同过滤算法[J].计算机工程与应用, 2015, 51(10):128-131. doi: 10.3778/j.issn.1002-8331.1405-0379 [9] LEMIRE D, MACLACHLAN A. Slope one predictors for online rating-based collaborative filtering[C]//Proceedings of the SIAM Data Mining (SDM'05). 2005:21-23. [10] LIU F, LEE H J. Use of social network information to enhance collaborative filtering performance[J]. Expert Systems with Applications, 2010, 37(7):4772-4778. doi: 10.1016/j.eswa.2009.12.061 [11] YUAN W, GUAN D, LEE Y, et al. Improved trust-aware recommender system using small worldness of trust networks[J]. Knowledge Based Systems, 2010, 23(3):232-238. doi: 10.1016/j.knosys.2009.12.004 [12] DING L, STEIL D, DIXON B, et al. A relation context oriented approach to identify strong ties in social networks[J]. Knowledge-Based Systems, 2011, 24(8):1187-1195. doi: 10.1016/j.knosys.2011.05.006 [13] 邓晓懿, 金淳, 韩庆平, 等.基于情境聚类和用户评级的协同过滤推荐模型[J].系统工程理论与实践, 2013, 33(11):2945-2953. doi: 10.12011/1000-6788(2013)11-2945 [14] MA H, KING I, LYU M R. Effective missing data prediction for collaborative filtering[C]//Sigir Proceedings of Annual International ACM Sigir Conference on Research & Development 2007:39-46. [15] LI Y M, LAI C Y, CHEN C W. Discovering influencers for marketing in the blogosphere[J]. Information Sciences, 2011, 181(23):5143-5157. doi: 10.1016/j.ins.2011.07.023 [16] YUAN W W, GUAN D H, LEE Y K, et al. Improved trust-aware recommender system using small-worldness of trust networks[J]. Knowledge Based Systems, 2010, 23(3):232-238. doi: 10.1016/j.knosys.2009.12.004 [17] ARASU A, CHO J, GARCIA-MOLINA H, et al. Searching the Web[J]. ACM Transactions on Internet Technology, 2001(1):2-43. -
点击查看大图
图(4) / 表(5)
计量
- 文章访问数: 241
- HTML全文浏览量: 158
- PDF下载量: 367
- 被引次数: 0