基于元图归一化相似性度量的实体推荐

doi:10.6040/j.issn.1672-3961.0.2019.304

摘要/Abstract

摘要：

基于异构信息网络(heterogeneous information networks, HIN)中元图的良好表征特性,提出一种结合隐式反馈和PathSim(meta path-based similarity)的归一化相似性度量(normalized similarity measure of meta graph, NSMG),以解决对异构信息网络中高度可见实体的偏好问题。针对Yelp和Amazon数据集构建Yelp-HIN(heterogeneous information networks in Yelp)和Amazon-HIN(heterogeneous information networks in Amazon),定义不同类型的元图及归一化相似度量,使用矩阵分解和因子分解机来组合计算不同元图上的相似性。试验结果表明,基于NSMG的方法在非常稀疏的数据集上性能表现优于常用的HIN实体推荐方法。

关键词: 异构信息网络, 元图, 归一化相似性度量, 实体推荐, 矩阵分解, 因子分解机

Abstract:

Based on the promising result of meta graph in heterogeneous information networks (HIN), normalized similarity measure of meta graph (NSMG) was proposed which combined implicit feedback matrix and PathSim(meta path-based similarity) to solve the problem of preference for large degree entities. Yelp-HIN(heterogeneous information networks in Yelp) and Amazon-HIN(heterogeneous information networks in Amazon) were constructed based on Yelp and Amazon datasets. Different types of meta graphs and normalized similarity measures were defined. Matrix decomposition and factorization machine were used to combine the similarities on different meta graphs. The experimental results showed that the proposed method based on normalization similarity measure of meta graphs performed better than the commonly used entity recommendation method in HIN on very sparse data sets.

Key words: heterogeneous information networks, meta graph, normalized similarity measure, entity recommendation, matrix decomposition, factorization machine

中图分类号:

TP311

张文凯,禹可,吴晓非. 基于元图归一化相似性度量的实体推荐[J]. 山东大学学报 (工学版), 2020, 50(2): 66-75.

Wenkai ZHANG,Ke YU,Xiaofei WU. Entity recommendation based on normalized similarity measure of meta graph in heterogeneous information network[J]. Journal of Shandong University(Engineering Science), 2020, 50(2): 66-75.

图/表 13

图1

图2

图3

表1

图4

图5

图6

表2

表3

表4

图7

图8

图9

参考文献 25

1	ZHAO H, YAO Q M, LI J D, et al. Meta-graph based recommendation fusion over heterogeneous information networks[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Halifax, Canada: ACM, 2017: 635-644.
2	SHI C , LI Y T , ZHANG J W , et al. A survey of heterogeneous information network analysis[J]. IEEE Transactions on Knowledge and Data Engineering, 2017, 29 (1): 17- 37. doi: 10.1109/TKDE.2016.2598561
3	JEH G, WIDOM J. Scaling personalized web search[C]//Proceedings of the 12th international conference on World Wide Web. Budapest, Hungary: ACM, 2003: 271-279.
4	郑玉艳, 田莹, 石川. 一种元路径下基于频繁模式的实体集扩展方法[J]. 软件学报, 2018, 29 (10): 2915- 2930.
	ZHENG Yuyan , TIAN ying , SHI Chuan . Method of entity set expansion based on frequent pattern under meta path[J]. Journal of Software, 2018, 29 (10): 2915- 2930.
5	黄立威, 李德毅, 马于涛, 等. 一种基于元路径的异质信息网络链路预测模型[J]. 计算机学报, 2014, 37 (4): 848- 858.
	HUANG Liwei , LI Deyi , MA Yutao , et al. A meta path-based link prediction model for heterogeneous information networks[J]. Chinese Journal of Computers, 2014, 37 (4): 848- 858.
6	盛权为, 汪一百, 高阳. 一种改进的异构链路协同预测算法研究[J]. 计算机工程与应用, 2017, 53 (15): 155- 163.
	SHENG Quanwei , WANG Yibai , GAO Yang . Research on improved algorithm for collaborative prediction of heterogeneous links[J]. Computer Engineering and Applications, 2017, 53 (15): 155- 163.
7	SUN Y Z , HAN J W , YAN X F , et al. Pathsim: meta path-based top-k similarity search in heterogeneous information networks[J]. Proceedings of the VLDB Endowment, 2011, 4 (11): 992- 1003.
8	SHI C , KONG X N , HUANG Y , et al. Hetesim: a general framework for relevance measure in heterogeneous networks[J]. IEEE Transactions on Knowledge and Data Engineering, 2014, 26 (10): 2479- 2492. doi: 10.1109/TKDE.2013.2297920
9	HUANG Z P, ZHENG Y D, CHENG R, et al. Meta structure: computing relevance in large heterogeneous information networks[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, USA: ACM, 2016: 1595-1604.
10	YU X, REN X, SUN Y Z, et al. Recommendation in heterogeneous information networks with implicit user feedback[C]//Proceedings of the 7th ACM conference on Recommender systems. Hong Kong, China: ACM, 2013: 347-350.
11	YU X, REN X, SUN Y Z, et al. Personalized entity recommendation: a heterogeneous information network approach[C]//Proceedings of the 7th ACM international conference on Web search and data mining. New York, USA: ACM, 2014: 283-292.
12	SHI C , LIU J , ZHUANG F Z , et al. Integrating heterogeneous information via flexible regularization framework for recommendation[J]. Knowledge and Information Systems, 2016, 49 (3): 835- 859. doi: 10.1007/s10115-016-0925-0
13	ZHENG J, LIU J, SHI C, et al. Dual similarity regularization for recommendation[C]// Pacific-Asia Conference on Knowledge Discovery and Data Mining. Auckland, New Zealand: Springer, 2016: 542-554.
14	JAMALI M, LAKSHMANAN L. HeteroMF: recommendation in heterogeneous information networks using context dependent factor models[C]// Proceedings of the 22nd International Conference on World Wide Web. Rio de Janeiro, Brazil: ACM, 2013: 643-654.
15	XIE F, CHEN L, YE Y, et al. A weighted meta-graph based approach for mobile application recommendation on heterogeneous information networks[C]//International Conference on Service-Oriented Computing. Hangzhou, China: Springer, 2018: 404-420.
16	ZHENG J, LIU J, SHI C, et al. Dual similarity regularization for recommendation[C]// Pacific-Asia Conference on Knowledge Discovery and Data Mining. Auckland, New Zealand: Springer, 2016: 542-554.
17	KOREN Y , BELL R , VOLINSKY C . Matrix factorization techniques for recommender systems[J]. Computer, 2009, 42 (8): 30- 37. doi: 10.1109/MC.2009.263
18	SHI C, ZHOU C, KONG X N, et al. Heterecom: a semantic-based recommendation system in heterogeneous networks[C]//Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Beijing, China: ACM, 2012: 1552-1555.
19	SHI C, ZHANG Z Q, LUO P, et al. Semantic path based personalized recommendation on weighted heterogeneous information networks[C]//Proceedings of the 24th ACM International on Conference on Information and Knowledge Management. Melbourne, Australia: ACM, 2015: 453-462.
20	BURKE R, VAHEDIAN F, MOBASHER B. Hybrid recommendation in heterogeneous networks[C]//International Conference on User Modeling, Adaptation, and Personalization. Aalborg, Denmark: Springer, 2014: 49-60.
21	王瑜, 武延军, 吴敬征, 等. 基于异构网络面向多标签系统的推荐模型研究[J]. 软件学报, 2017, 28 (10): 2611- 2624.
	WANG Yu , WU Yanjun , WU Jingzheng , et al. Multi-Dimensional tag recommender model via heterogeneous networks[J]. Journal of Software(in Chinese), 2017, 28 (10): 2611- 2624.
22	王永, 邓永恒, 李晓光. 考虑非对称用户偏好的推荐算法[J]. 计算机工程与应用, 2018, 54 (23): 1- 6. doi: 10.3778/j.issn.1002-8331.1809-0322
	WANG Yong , DENG Yongheng , LI Xiaoguang . Asymmetric recommendation algorithm based on user preference[J]. Computer Engineering and Applications, 2018, 54 (23): 1- 6. doi: 10.3778/j.issn.1002-8331.1809-0322
23	戴琳, 孟祥武, 张玉洁, 等. 一种融合多种数据信息的餐馆推荐模型[J]. 软件学报, 2019, 30 (9): 2869- 2885.
	DAI Lin , MENG Xiangwu , ZHANG Yujie , et al. A restaurant recommendation model with multiple information fusion[J]. Journal of Software(in Chinese), 2019, 30 (9): 2869- 2885.
24	YUAN M , LIN Y . Model selection and estimation in regression with grouped variables[J]. Journal of the Royal Statistical Society: Series B:Statistical Methodology, 2006, 68 (1): 49- 67. doi: 10.1111/j.1467-9868.2005.00532.x
25	LI H, LIN Z. Accelerated proximal gradient methods for nonconvex programming[C]//Advances in Neural Information Processing Systems. Montreal, Canada: NIPS, 2015: 379-387.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

HIN	元路径
Yelp-HIN	M₃:用户-商家-分类-商家
	M₄:用户-商家-州-商家
	M₅:用户-商家-城市-商家
	M₆:用户-商家-星级-商家
	M₇:用户-商家-分类-商家-分类-商家
	M₈:用户-商家-州-商家-州-商家
	M₉:用户-评论-主题-评论-用户-商家
	M₁₀:用户-评论-商家-评论-用户-商家
Amazon-HIN	M₃:用户-产品-分类-产品
	M₄:用户-产品-品牌-产品
	M₅:用户-产品-分类-产品-分类-产品
	M₆:用户-产品-品牌-产品-品牌-产品
	M₇:用户-评论-主题-评论-用户-产品
	M₈:用户-评论-产品-评论-用户-产品

统计信息	用户数/个	商家/个	类别数/个	星级数/个	城市数/个	州数/个	品牌数/个	签到数/个	签到数:用户数	签到数:商家数
Yelp-50k	13 663	8 164	716	9	154	13		37 659	2.76	4.61
Amazon-50k	20 345	8 146	488				640	29 350	1.44	3.60

数据集	稠密度/%
Yelp-50k	0.037
Amazon-50k	0.017
IMDb-MovieLens-100k	6.988
Yelp^[19]	0.086
Douban	0.630

方法	Yelp-50k		Amazon-50k
方法	RMSE	NSMG提升百分比/%	RMSE	NSMG提升百分比/%
RSVD	2.599 6	52.6	2.639 5	55.5
HeteRec	1.732 0	28.9	1.936 0	39.3
FMG	1.247 3	1.2	1.185 0	1.3
NSMG	1.231 8		1.174 1

[1]	段圣宇,吴伊宁,赛高乐. 一种面向矩阵分解模型的推荐系统训练加速方法[J]. 山东大学学报 (工学版), 2025, 55(1): 24-29.
[2]	王冰,马文明,武聪,郝昱猛. 融合信任相似度的偏置概率矩阵分解算法[J]. 山东大学学报 (工学版), 2022, 52(4): 110-117.
[3]	林晓炜,陈黎飞. 结构扩展的非负矩阵分解社区发现算法[J]. 山东大学学报 (工学版), 2021, 51(2): 57-64.
[4]	黄丹,王志海,刘海洋. 一种局部协同过滤的排名推荐算法[J]. 山东大学学报(工学版), 2016, 46(5): 29-36.
[5]	庞俊涛, 张晖, 杨春明, 李波, 赵旭剑. 基于概率矩阵分解的多指标协同过滤算法[J]. 山东大学学报(工学版), 2016, 46(3): 65-73.
[6]	李新玉, 徐桂云,任世锦,杨茂云. 基于鉴别流形的不相关稀疏投影非负矩阵分解[J]. 山东大学学报 (工学版), 2015, 45(5): 1-12.
[7]	李国栋,赵威,田国会*,薛英花. 一种基于旋转矩阵分解的视觉伺服控制算法[J]. 山东大学学报(工学版), 2012, 42(1): 45-50.