异质网络中基于节点影响力的相似度度量方法

doi:10.3969/j.issn.0372-2112.2019.09.016

电子学报 ›› 2019, Vol. 47 ›› Issue (9): 1929-1936.DOI: 10.3969/j.issn.0372-2112.2019.09.016

异质网络中基于节点影响力的相似度度量方法

刘露^1,2,3,4, 胡封晔⁴, 牛亮⁵, 彭涛^1,2,3

1. 吉林大学软件学院, 吉林长春 130012;
2. 符号计算与知识工程教育部重点实验室(吉林大学), 吉林长春 130012;
3. 吉林大学计算机科学与技术学院, 吉林长春 130012;
4. 吉林大学通信工程学院, 吉林长春 130012;
5. 吉林大学第一医院, 吉林长春 130012

收稿日期:2018-06-15 修回日期:2019-04-17 出版日期:2019-09-25
- 通讯作者:
- 彭涛
- 作者简介:
- 刘露女,1989年生于辽宁大连,博士、讲师.主要研究方向为数据挖掘、机器学习、自然语言处理、异质信息网络挖掘.E-mail:liulu@jlu.edu.cn;胡封晔 男,1974年生于河南原阳,博士、教授.主要研究方向为信号处理,无线体域网、认知无线电网络、空时通信和无线定位等.E-mail:fufy@jlu.edu.cn;牛亮女,1980年生于吉林长春,硕士,主要研究方向为社会网络.E-mail:379879995@qq.com
- 基金资助:
- 国家自然科学基金 (No.61872163，No.61806084）; 中国博士后科研基金项目 (No.2018M631872）; 吉林省教育厅项目 (No.JJKH20190160KJ）; 吉林省科技厅重点科技研发项目 (No.20180201044GX）

Node Influence Based Similarity Measure Method in Heterogeneous Network

LIU Lu^1,2,3,4, HU Feng-ye⁴, NIU Liang⁵, PENG Tao^1,2,3

1. College of Software, Jilin University, Changchun, Jilin 130012, China;
2. Key Laboratory of Symbolic Computation and Knowledge Engineering(Jilin University), Ministry of Education, Changchun, Jilin 130012, China;
3. College of Computer Science and Technology, Jilin University, Changchun, Jilin 130012, China;
4. College of Communication Engineering, Jilin University, Changchun, Jilin 130012, China;
5. The First Hospital of Jilin University, Changchun, Jilin 130012, China

Received:2018-06-15 Revised:2019-04-17 Online:2019-09-25 Published:2019-09-25
- Corresponding author:
- PENG Tao
- Supported by:
- National Natural Science Foundation of China (No.61872163, No.61806084); Program of Post-doctoral Research Fund of China (No.2018M631872); Program of Department of Education of Jilin Province (No.JJKH20190160KJ); Key Research and Development Project of Science and Technology Department of Jilin Province (No.20180201044GX)

摘要/Abstract

摘要： 异质网络相似度学习，即分析两个不同类型对象间的相关程度.不同类型对象在异质网络中的重要程度不同，它们在相似度学习过程中的发挥的作用也不同.针对异质网络，提出了一种基于节点影响力的相似度度量方法NISim，该模型既考虑了网络中的链接结构，也保留了网络中的语义信息，同时区分不同类型节点对异质网络的作用.在异质信息网络环境下，通过启发式规则区分并量化不同类型节点的影响力权值，并结合网络链接结构和节点间语义关系，解决了提高相似度学习准确性的问题.实验结果表明，该方法能够有效地对异质信息网络不同类型节点进行相似度度量，可以应用在网络搜索、推荐系统以及知识图谱构建等不同领域.

关键词: 数据挖掘, 异质网络, 推荐系统, 知识图谱, 网络搜索, 节点影响力, 链接结构, 语义关系

Abstract: Heterogeneous network similarity learning is to analyze the degree of correlation between two different types of objects. Different types of objects have different degrees of importance in heterogeneous networks, and play different roles in the similarity learning process.This paper proposes a node influence based similarity measure method (NISim) heterogeneous information network. This method not only considers the link structure in network but also keeps the semantic information in heterogeneous networks. Also, this method distinguishes the effect to heterogeneous network brought by different types of nodes. In heterogeneous network, the heuristic rules are used to distinguish and quantify the influence weight of different types of nodes. In addition, the link structure in network and the semantic relationship are combined to solve the problem of improving similarity learning accuracy. Experimental results show that this method can measure the similarity between different types of nodes effectively. It can be applied in different fields such as network search, recommendation system and knowledge graph construction and so on.

Key words: data mining, heterogeneous network, recommended system, knowledge graph, network search, node influence, link structure, semantic relationship

中图分类号:

TP391

刘露, 胡封晔, 牛亮, 等. 异质网络中基于节点影响力的相似度度量方法[J]. 电子学报, 2019, 47(9): 1929-1936.

LIU Lu, HU Feng-ye, NIU Liang, et al. Node Influence Based Similarity Measure Method in Heterogeneous Network[J]. Acta Electronica Sinica, 2019, 47(9): 1929-1936.

参考文献

[1] WEI Y,SONG Y Q,ZHEN Y,LIU B,YANG Q.Heterogeneous translated hashing:A scalable solution towards multi-modal similarity search[J].ACM Transactions on Knowledge Discovery from Data,2016,10(4):article no.36.
[2] NGUYEN P,WANG J,HILARIO M,KALOUSIS A.Learning heterogeneous similarity measures for hybrid recommendations in meta-mining[A].Proceedings of the 12th IEEE International Conference on Data Mining[C].Washington,DC,USA:IEEE Computer Society,2012.1026-1031.
[3] WU J,SHEN H,LI Y D,XIAO Z B,LU M Y,WANG C L.Learning a hybrid similarity measure for image retrieval[J].Pattern Recognition,2013,46(11):2927-2939.
[4] LIN Y K,LIU Z Y,SUN M S,LIU Y,ZHU X.Learning entity and relation embeddings for knowledge graph completion[A].Proceedings of the 29th AAAI Conference on Artificial Intelligence[C].Austin,Texas,USA:AAAI Press,2015.2181-2187.
[5] WANG C G,SONG Y Q,LI H R,ZHANG M,HAN J W.KnowSim:A document similarity measure on structured heterogeneous information networks[A].Proceedings of IEEE International Conference on Data Mining[C].Washington,DC,USA:IEEE Computer Society,2015.1015-1020.
[6] MEHDI O,IBRAHIM H,AFFENDEY L.An approach for instance based schema matching with Google similarity and regular expression[J].International Arab Journal of Information Technology,2017,14(5):755-763.
[7] JUNEJO K N,KARIM A,HASSAN M T,JEON M.Terms-based discriminative information space for robust text classification[J].Information Sciences,2016,372:518-538.
[8] TRIPATHY A,ANAND A,RATH S.K.Document-level sentiment classification using hybrid machine learning approach[J].Knowledge and Information Systems,2017,53(3):805-831.
[9] ZHANG Y L,LI C P,XIE C W,CHEN H.Accuracy estimation of link-based similarity measures and its application[J].Frontiers of Computer Science,2016,10(1):113-123.
[10] HAMEDANI M R,KIM S W.JacSim:An accurate and efficient link-based similarity measure in graphs[J].Information Sciences,2017,414:203-224.
[11] LIU H P,JIN C Q,YANG B,ZHOU A Y.Finding top-k shortest paths with diversity[J].IEEE Transactions on Knowledge and Data Engineering,2018,30(3):488-502.
[12] SUN Y Z,HAN J W,YAN X F,YU P S,WU T Y.PathSim:meta path-based top-k similarity search in heterogeneous information networks[A].Proceedings of the VLDB Endowment[C].Seattle,Washington,USA:ACM Press,2011.4(11):992-1003.
[13] GAO X J,MU T T,GOULERMAS J Y,WANG M.Topic driven multimodal similarity learning with multi-view voted convolutional features[J].Pattern Recognition,2018,75:223-234.
[14] XIONG Y,ZHU Y Y,YU P S.Top-k similarity join in heterogeneous information networks[J].IEEE Transactions on Knowledge and Data Engineering,2015,27(6):1710-1723.
[15] ZHANG M X,HU H,HE Z Y,WANG W.Top-k similarity search in heterogeneous information networks with x-star network schema[J].Expert Systems with Applications,2015,42(2):699-712.
[16] SHAMS B,HARATIZADEH S.IteRank:A iterative network-oriented approach to neighbor-based collaborative ranking[J].Knowledge-based Systems,2017,128:102-114.
[17] TAO W B,YU M H,LI G L.Efficient top-k simrank-based similarity join[A].Proceedings of the VLDB Endowment[C].Hangzhou,China:ACM Press,2014.8(3):317-328.
[18] SHI C,KONG X,HUANG Y,YU P S,WU B.HeteSim:A general framework for relevance measure in heterogeneous information networks[J].IEEE Transactions on Knowledge and Data Engineering,2014,26(10):2479-2492.
[19] JAIN Y K,BHANDARE S K.Min max normalization based data perturbation method for privacy protection[J].International Journal of Computer & Communication Technology,2011,2(8):45-50.
[20] 刘露.异质信息网络中离群点检测方法研究[D].长春:吉林大学,2017. LIULu.Research on Outlier Detection Methods in Heterogeneous Networks[D].Changchun:Jilin University,2017(in Chinese)
[21] 王岩,彭涛,韩佳育,等.一种基于密度的分布式聚类方法[J].软件学报,2017,28(11):2836-2850. WANGYan,PENG Tao,HAN Jia-yu,et al.Density-based distributed clustering method[J].Journal of Software,2017,28(11):2836-2850.(in Chinese)
[22] HUANG X H,YE Y M,XIONG L Y,WANG S K,YANG X F.Clustering time-stamped data using multiple nonnegative matrices factorization[J].Knowledge-based Systems,2016,114:88-98.
[23] TANG J,ZHANG J,YAO L M,LI J Z,ZHANG L,SU Z.ArnetMiner:Extraction and mining of academic social networks[A].Proceedings of the 14th ACM SIGKDD International Conference of Knowledge Discovery and Data Mining[C].Las Vegas,Nevada,USA:ACM Press,2008.990-998.
[24] MENG X F,SHI C,LI Y T,ZHANG L,WU B.Relevance measure in large-scale heterogeneous networks[A].Asia-Pacific Web Conference[C].Germany:Springer,2014.636-643.

异质网络中基于节点影响力的相似度度量方法

Node Influence Based Similarity Measure Method in Heterogeneous Network

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

[1]	曹阳, 高旻, 余俊良, 范琪琳, 荣文戈, 文俊浩. 基于双图混合随机游走的社会化推荐模型[J]. 电子学报, 2023, 51(2): 286-296.
[2]	皮德常, 吴致远, 曹建军. 基于知识图谱表示学习的谣言早期检测方法[J]. 电子学报, 2023, 51(2): 385-395.
[3]	张少波, 原刘杰, 毛新军, 朱更明. 基于本地差分隐私的K-modes聚类数据隐私保护方法[J]. 电子学报, 2022, 50(9): 2181-2188.
[4]	沈鹏飞, 徐臻, 王英. 基于嵌套生成对抗学习的网络嵌入[J]. 电子学报, 2022, 50(9): 2155-2163.
[5]	刘群, 谭洪胜, 张优敏, 王国胤. 基于元路径的动态异质网络表示学习[J]. 电子学报, 2022, 50(8): 1830-1839.
[6]	张帅, 高旻, 文俊浩, 熊庆宇, 唐旭. 基于自监督学习的去流行度偏差推荐方法[J]. 电子学报, 2022, 50(10): 2361-2371.
[7]	王森, 王煜, 宁德军. 复杂软件系统健康状态智能感知与诊断模型[J]. 电子学报, 2021, 49(9): 1799-1808.
[8]	高留杰, 赵文, 张君福, 姜波. G2S:基于语义块的知识图谱问答语义解析[J]. 电子学报, 2021, 49(6): 1132-1141.
[9]	李斌, 陈定山, 孙小兵, 薄莉莉. 面向缺陷知识的多特征匹配搜索算法[J]. 电子学报, 2021, 49(4): 661-664.
[10]	郑孝遥, 罗永龙, 汪祥舜, 孙丽萍, 陈付龙, 胡桂银, 汪小寒. 基于位置服务的分布式差分隐私推荐方法研究[J]. 电子学报, 2021, 49(1): 99-110.
[11]	郭景峰, 朱晓松, 李爽. 直播电视推荐系统的评分预测算法研究[J]. 电子学报, 2020, 48(9): 1735-1740.
[12]	黄名选, 蒋曹清. 基于项权值排序挖掘的跨语言查询扩展[J]. 电子学报, 2020, 48(3): 568-576.
[13]	吴玉佳, 李晶, 宋成芳, 常军. 基于高效用神经网络的文本分类方法[J]. 电子学报, 2020, 48(2): 279-284.
[14]	高鹏, 陈智华. 一种基于拓扑信息的预测疾病相关的MicroRNAs方法[J]. 电子学报, 2020, 48(2): 333-340.
[15]	方晨, 郭渊博, 王娜, 甄帅辉, 唐国栋. 基于生成对抗网络的差分隐私数据发布方法[J]. 电子学报, 2020, 48(10): 1983-1992.