基于Attention深度随机森林的社区演化事件预测

doi:10.3969/j.issn.0372-2112.2019.10.005

电子学报 ›› 2019, Vol. 47 ›› Issue (10): 2050-2060.DOI: 10.3969/j.issn.0372-2112.2019.10.005

基于Attention深度随机森林的社区演化事件预测

潘剑飞^1,2, 曹燕¹, 董一鸿¹, 陈华辉¹, 钱江波¹

1. 宁波大学信息科学与工程学院, 浙江宁波 315211;
2. 北京百度在线科技有限公司, 北京 100084

收稿日期:2018-06-11 修回日期:2019-03-09 出版日期:2019-10-25
- 通讯作者:
- 董一鸿
- 作者简介:
- 潘剑飞 男,1991年生.CCF学生会员,2018年获宁波大学计算机技术专业硕士学位,现为百度公司算法工程师,主要研究方向为大数据、数据挖掘;曹燕女,1993年生.CCF学生会员,宁波大学信息科学与工程学院硕士,主要研究方向为大数据、数据挖掘;陈华辉 男,1964年出生于浙江鄞州.博士,CCF会员,宁波大学教授,主要研究方向为数据库技术、流数据处理等.E-mail:chenhuahui@nbu.edu.cn;钱江波 男,1974年出生于浙江宁波.博士,CCF高级会员,宁波大学教授,主要研究方向为数据库技术、流数据处理、多维数据索引技术等.E-mail:qianjiangbo@nbu.edu.cn
- 基金资助:
- 国家自然科学基金 (No.61472194，No.61572266）; 浙江省自然科学基金 (No.LY16F020003）; 宁波市自然科学基金 (No.2017A610114）

The Community Evolution Event Prediction Based on Attention Deep Random Forest

PAN Jian-fei^1,2, CAO Yan¹, DONG Yi-hong¹, CHEN Hua-hui¹, QIAN Jiang-bo¹

1. Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, Zhejiang 315211, China;
2. Baidu Online Technology Co., Ltd., Beijing 100084, China

Received:2018-06-11 Revised:2019-03-09 Online:2019-10-25 Published:2019-10-25
- Corresponding author:
- DONG Yi-hong
- Supported by:
- National Natural Science Foundation of China (No.61472194, No.61572266); National Natural Science Foundation of Zhejiang Province, China (No.LY16F020003); Ningbo Natural Science Fund (No.2017A610114)

摘要/Abstract

摘要： 在网络结构不断变化的同时，社区结构也随之演化.社区结构在不同时间片的变化可定义为四种不同的演化事件：持续、分离、融合和消失.本文运用网络表示学习的方法，对网络进行图嵌入编码映射到低维向量空间中，研究动态社区演化事件的预测.特征方面，在传统的社区内部属性特征、时间片间属性特性变化和前段时间片的社区演化事件的特征维度的基础上，引入潜在结构特征表征四种演化事件，运用随机游走和Softmax思想获取潜在的结构特征；模型方面，引入深度随机森林的策略，同时采用attention机制、蒙特卡洛特征采样策略进行特征融合和特征训练，克服了已有算法仅获取局部结构特征的缺陷.实验在DBLP、FACEBOOK和Bitcoin数据集上，对比SVM、XGBOOST和RIDGE模型训练，证实了新提出的算法模型对最终预测准确率有很大的提升.

关键词: 社区演化, 图嵌入, 网络表示学习, 深度随机森林, attention机制

Abstract: The internal community structure is evolving with the change of network structure. These changes in different time slices can be defined as four different evolutionary events: survive, split, fusion and disappearance. In this paper, the network representation learning method is used to map the graph embedding of the network into the low-dimensional vector space to study the prediction of dynamic community evolution events. In the features, based on the features of community internal attributes, the change of temporal attributes,and the previous community evolution events, the potential structure characteristics of the four evolutionary events are introduced and obtained by using random walk and Softmax. In the model,the strategy of deep random forest is proposed. Feature fusion and feature training are carried out by using the attention mechanism and Monte Carlo feature sampling strategy, which overcomes the shortcomings of the existing algorithms that only acquire local structural features. Finally, by comparing SVM, XGBOOST, RIDGE model training in the DBLP, FACEBOOK and Bitcoin datasets, it is confirmed that the embedding feature of community structure and the attention deep random forest model improvement have greatly improved the accuracy of final prediction.

Key words: community evolution, graph embedding, network representation learning, deep random forests, attention mechanism

中图分类号:

TP311

潘剑飞, 曹燕, 董一鸿, 等. 基于Attention深度随机森林的社区演化事件预测[J]. 电子学报, 2019, 47(10): 2050-2060.

PAN Jian-fei, CAO Yan, DONG Yi-hong, et al. The Community Evolution Event Prediction Based on Attention Deep Random Forest[J]. Acta Electronica Sinica, 2019, 47(10): 2050-2060.

参考文献

[1] YUANZHUO W,XIAOLONG J,XUEQI C.Network big data:Present and future[J].Chinese Journal of Computers,2013,36(6):1125-1138.
[2] XIE J,SZYMANSKI B K,LIU X.Slpa:Uncovering overlapping communities in social networks via a speaker-listener interaction dynamic process[A].IEEE 11th International Conference on Data Mining Workshops[C].US:IEEE,2011.344-349.
[3] 马慧芳,谢蒙,何廷年,等.基于核心标签的可重叠微博网络社区划分方法[J].电子学报,2017,45(4):769-776. MA Hui-fang,Xie Meng,HE Yan-nian,et al.An overlapping microblog community detection algorithm via core tags[J].Acta Electronica Sinica,2017,45(4):769-776.(in Chinese)
[4] 王诗懿,董一鸿,李志超,等.大规模复杂网络下重叠社区的识别[J].电子学报,2015,43(8):1575-1582. WANG Shi-yi,DONG Yi-hong,LI Zhi-chao,et al.The identification of over lapping communities in large-scale complex networks[J].Acta Electronica Sinica,2015,43(8):1575-1582.(in Chinese)
[5] BRODKA P,SAGANOWSKI S,KAZIENKO P.GED:the method for group evolution discovery in social networks[J].Social Network Analysis and Mining,2013,3(1):1-14.
[6] WU J,YANG H.Linear regression-based efficient SVM learning for large-scale classification[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(10):2357-2369.
[7] CHEN T,HE T,BENESTY M,et al.Xgboost:Extreme Gradient Boosting[OL].https://github.com/dmlc/xgboost.Accessed:2017-10-18.
[8] MCDONALD G C.Ridge regression[J].Wiley Interdisciplinary Reviews:Computational Statistics,2009,1(1):93-100.
[9] BACKSTROM L,HUTTENLOCHER D,KLEINBERG J,et al.Group formation in large social networks:membership,growth,and evolution[A].Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining[C].US:ACM,2006.44-54.
[10] FENG X,ZHAO J C,XU K.Link prediction in complex networks:A clustering perspective[J].The European Physical Journal B,2012,85(1):3.
[11] GOLDBERG M,MAGDON-ISMAIL M,NAMBIRAJAN S,et al.Tracking and predicting evolution of social communities[A].IEEE Third International Conference on Privacy,Security,Risk and Trust and IEEE Third International Conference on Social Computing[C].US:IEEE,2011.780-783.
[12] AU W H,CHAN K C C,YAO X.A novel evolutionary data mining algorithm with applications to churn prediction[J].IEEE Transactions on Evolutionary Computation,2003,7(6):532-545.
[13] BRODKA P,KAZIENKO P,KOLOSZCZYK B.Predicting group evolution in the social network[A].International Conference on Social Informatics[C].Berlin,Heidelberg:Springer,2012.54-67.
[14] ZHOU Z H,FENG J.Deep forest:Towards an alternative to deep neural networks[J].Arxiv Preprint,2017,arXiv:1702.08835.
[15] ANDERSEN R,CHUNG F,LANG K.Local graph partitioning using pagerank vectors[A].The 47th Annual IEEE Symposium on Foundations of Computer Science (FOCS'06)[C].US:IEEE,2006.475-486.

基于Attention深度随机森林的社区演化事件预测

The Community Evolution Event Prediction Based on Attention Deep Random Forest

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

[1]	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.
[2]	孙全明, 曲志坚, 任崇广. 基于粒子群优化和LightGBM的情景感知多式联运推荐[J]. 电子学报, 2021, 49(5): 894-903.
[3]	黄鸿, 王丽华, 石光耀. 面向高光谱遥感影像分类的监督多流形鉴别嵌入方法[J]. 电子学报, 2020, 48(6): 1099-1107.
[4]	程晓涛, 吉立新, 尹赢, 黄瑞阳. 基于D-S证据理论的网络表示融合方法[J]. 电子学报, 2020, 48(5): 854-860.
[5]	祁志卫, 王笳辉, 岳昆, 乔少杰, 李劲. 图嵌入方法与应用:研究综述[J]. 电子学报, 2020, 48(4): 808-818.
[6]	尹赢, 张建朋, 吉立新, 李治成. 基于霍克斯点过程的动态网络表示学习方法[J]. 电子学报, 2020, 48(11): 2154-2161.
[7]	曹燕, 董一鸿, 邬少清, 陈华辉, 钱江波, 潘善亮. 动态网络表示学习研究进展[J]. 电子学报, 2020, 48(10): 2047-2059.
[8]	王立志;黄鸿;冯海亮. 基于SSMFA与kNNS算法的高光谱遥感影像分类[J]. 电子学报, 2012, 40(4): 780-787.
[9]	陈国明;印鉴;周端宁;张东. 一种新的隐写分析方法:IKLDA[J]. 电子学报, 2009, 37(8): 1762-1767.