淡江大學覺生紀念圖書館 (TKU Library)
進階搜尋


下載電子全文限經由淡江IP使用) 
系統識別號 U0002-2907202010415500
中文論文名稱 以生成對抗網路為基礎的動態社群網路預測
英文論文名稱 Dynamic Social Network Prediction Based on Generative Adversarial Network
校院名稱 淡江大學
系所名稱(中) 資訊工程學系全英語碩士班
系所名稱(英) Master’s Program, Department of Computer Science and Information Engineering (English-taught program
學年度 108
學期 2
出版年 109
研究生中文姓名 劉俐伶
研究生英文姓名 Li-Ling Liu
學號 607780011
學位類別 碩士
語文別 英文
口試日期 2020-07-14
論文頁數 30頁
口試委員 指導教授-王英宏
委員-惠霖
委員-陳以錚
中文關鍵字 動態網路  社群網路連結預測  長短期記憶  生成對抗網路 
英文關鍵字 Dynamic Network  Social Network Link Prediction  Long Short-Term Memory(LSTM)  Generative Adversarial Network(GAN) 
學科別分類 學科別應用科學資訊工程
中文摘要 隨著網路媒體越來越發達,在虛擬世界中,人與人之間的關係已然形成一種社群網路,而這些關係會隨著時間有所改變,稱為動態社群網路。為了將網路作為圖形處理並考慮其時間特性,我們設置時間點作為固定間隔,並將一系列快照劃分為動態社群網路,以觀察節點的連結趨勢並預測哪些節點將出現或消失在圖中。 此過程稱為「動態網路連結預測」(Dynamic Network Link Prediction ,DNLP), 這意味著我們可以根據其歷史行為來推斷將與目標用戶建立關係的特定人群。我們提出了一種新的基於生成對抗網路(Generative Adversarial Network , GAN)模型,叫Soc-GAN,結合長短期記憶(Long Short-Term Memory, LSTM),用於動態網路的連結預測,並能處理長期預測任務,以捕獲序列之間的向量相關性,再加以分類、判斷生成的預測快照是否與真實資料相似。 同時,它也具有更穩健的能力來預測將在下一個網路圖中將出現或消失的連結。
英文摘要 With the development of online media, in the online virtual world, the relationship between people has formed a kind of social network, and these relationships will change over time, called dynamic social network. In order to deal the network as a graph and consider its time characteristics, we set the time point as a fixed interval and divide a series of snapshots into a dynamic social network to observe the connection trend of nodes and predict which nodes will appear or disappear in the graph. This process is called "Dynamic Network Link Prediction" (DNLP), which means that we can infer a specific group of people who will establish a relationship with the target user based on their historical behavior. We propose a new model based on Generative Adversarial Network (GAN), called Soc-GAN, combined with Long Short-Term Memory (LSTM), to do the link prediction of dynamic social networks and deal with the long-term prediction tasks for capture the relevance of vectors between the sequences and classify, distinguish whether the generated prediction snapshot is similar to real data. At the same time, it also has a more robust ability to predict the links that are going to appear or disappear in the next network graphs.
論文目次 Table of Contents
Chinese Abstract…………………………I
Abstract…………………………III
Table of Contents…………………………IV
List of Figures…………………………V
List of Tables…………………………VI
Chapter 1 Introduction…………………………1
Chapter 2 Related Work…………………………4
2.1 Social Network Prediction…………………………4
2.2 Generative Adversarial Network…………………………9
Chapter 3 Methodology…………………………12
3.1 Problem Definition…………………………12
3.1.1 Dynamic Network…………………………12
3.1.2 Dynamic Network Link Prediction…………………………12
3.2 Soc-GAN Framework…………………………13
3.2.1 Generator (LSTM)…………………………14
3.2.2 Discriminator…………………………16
3.3 Training Process…………………………16
Chapter 4 Experiments…………………………18
4.1 Dataset…………………………18
4.2 Baselines…………………………19
4.3 Evolution Metrics…………………………19
4.4 Experimental Results…………………………21
4.5 Parameter Setting…………………………23
Chapter 5 Conclusion…………………………26
Reference…………………………27

List of Figures
Figure 1: Dynamic social network prediction…………………………2
Figure 2: An illustration of network evolution…………………………13
Figure 3: The architecture of Soc-GAN…………………………14
Figure 4: The structure of the LSTM cell…………………………15
Figure 5: TPR vs. FPR at different classification thresholds…………………………21
Figure 6: AUC (Area under the ROC Curve)…………………………21
Figure 7: AUC performance of Soc-GAN models on different sliding window
size…………………………24
Figure 8: AUC performance of Soc-GAN models on different generator
layers…………………………24
Figure 9: AUC performance of Soc-GAN models on different learning rate…………………………25
Figure 10: AUC performance of Soc-GAN models on different discriminator
layers…………………………25
Figure 11: AUC performance of Soc-GAN models on different epochs…………………………25

List of Table
Table 1: The basic statistics of the three datasets…………………………18
Table 2: ROC confusion matrix…………………………20
Table 3: AUC performance of different models on three datasets…………………………23
參考文獻 Reference
[1] M.Abufouda, K. A. Zweig,” Interactions around social networks matter: predicting the social network from associated interaction networks,” IEEE/ACM, pp. 142–145, 2014.
[2] M.Arjovsky, S.Chintala, L.Bottou, “Wasserstein GAN,” arXiv ML, 2017.
[3] B.Chen, Y.Hua, Y.Yuan, Y.Jin,” Link Prediction on Directed Networks Based on AUC Optimization,” IEEE Access, Volume:6, pp.28122 - 28136, 2018.
[4] R. A. Rossi and N. K. Ahmed, “The network data repository with interactive graph analytics and visualization,” in Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015.
[5] X.Chen, Y.Duan, R.Houthooft, J.Schulman, I.Sutskever, P.Abbeel, “InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets,” arXiv CS, 2016.
[6] B.Cheng, X.Xu, Y.Zeng, J.Ren, S.Jung, “Pedestrian trajectory prediction via the Social-Grid LSTM model,” in the Journal of Engineering, pp.1468-1474, 2018.
[7] J.Chen, J.Zhang, X.Xu, C.Fu, D.Zhang, Q.Zhang,” E-LSTM-D: A Deep Learning Framework for Dynamic Network Link Prediction,” IEEE, 2019.
[8] R.Chen, Q.Hua, B.Wang, M.Zheng, W.Guan, X. Ji,” A Novel Social Recommendation Method Fusing User’s Social Status and Homophily Based on Matrix Factorization Techniques,”IEEE Access, Volume:7, pp.18783 - 18798, 2019.
[9] Y.Choi, M.Choi, M.Kim, J.Woo-Ha, S.Kim, J.Choo, “StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation,” arXiv CS, 2018.
[10] S.Dai, L.Li, Z.Li, ”Modeling Vehicle Interactions via Modified LSTM Models for Trajectory Prediction,” in IEEE Access ( Volume: 7 ), pp.38287-38296,2019.
[11] J.Donahue, P.Krähenbühl, T.Darrell,” Adversarial Feature Learning,” arXiv CS, 2017.
[12] S.Feng, D.Shen, T.Nie, Y.Kou, J.He, G.Yu,” Inferring Anchor Links Based on Social Network Structure,” IEEE Access, Volume:6, pp.17340 - 17353, 2018.
[13] A. Grover and J. Leskovec, “node2vec: Scalable feature learning for networks,” in Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2016, pp. 855–864.
[14] J.Gong, J.Tang, A. C. M. Fong,” ACTPred: Activity prediction in mobile social networks,” IEEE Access, Volume:19, pp.265-274, 2014.
[15] I.J. Goodfellow, J.Pouget-Abadie, M.Mirza, B.Xu, David, Warde-Farley, S.Ozair, A.Courville, Y.Bengio,” Generative Adversarial Networks,” NIPS'14, pp. 2672–2680, 2014.
[16] Z.Hao,” Link Prediction in Online Social Networks Based on the Unsupervised Marginalized Denoising Model.” IEEE Access, Volume:7, pp.54133 - 54143, 2019.
[17] “Haggle network dataset – KONECT,” Apr. 2017.
[18] S.Hochreiter, J.Schmidhuber,” Long Short-Term Memory,” Neural Computation, Volume 9 | Issue 8 | November 15, pp.1735-1780,1997.
[19] W.Hu, K.Kumar-Singh, F.Xiao, J.Han, C.Chuah, Y.Lee, “Who Will Share My Image?: Predicting the Content Diffusion Path in Online Social Network,” in the 11th ACM International Conference on Web Search and Data Mining, pp. 252-260, ACM, 2018.
[20] J.Huang, F.Nie, H.Huang, Y.Tu, Yu Lei,” Social trust prediction using heterogeneous networks,” ACM, Article No.:17, 2013.
[21] T. Li, B. Wang, Y. Jiang, Y. Zhang, and Y. Yan, “Restricted boltzmann machine-based approaches for link prediction in dynamic networks,” IEEE Access, 2018.
[22] T. Li, J. Zhang, S. Y. Philip, Y. Zhang, and Y. Yan, “Deep dynamic network embedding for link prediction,” IEEE Access, 2018.
[23] J.Liao, T.Liu, M.Liu, J.Wang, Y.Wang,” Multi-Context Integrated Deep Neural Network Model for Next Location Prediction,” IEEE Access, Volume:6, pp.21980 – 21990, 2018.
[24] D. Liben-Nowell and J. Kleinberg, “The link-prediction problem for social networks,” journal of the Association for Information Science and Technology, vol. 58, no. 7, pp. 1019–1031, 2007.
[25] Li, P., Li, J., Sun, F., & Wang, P. (2017, August). Short Text Emotion Analysis Based on Recurrent Neural Network. In Proceedings of the 6th International Conference on Information Engineering (p. 6). ACM.
[26] D.Li, D.Shen, Y.Kou, T.Nie,” Integrating Sign Prediction With Behavior Prediction for Signed Heterogeneous Information Networks,” IEEE Access, Volume:7, pp. 171357 - 171371, 2019.
[27] J.Li, J.Peng, S.Liu, X.Ji, Xing Li,” Link Prediction in Directed Networks Utilizing the Role of Reciprocal Links,” IEEE Access, Volume:8, pp.28668 - 28680, 2020.
[28] G.Liu, Y.Xu, Z.He, Y.Rao, J.Xia, L.Fan,” Deep Learning-Based Channel Prediction for Edge Computing Networks Toward Intelligent Connected Vehicles,” IEEE Volume:7, pp.114487 - 114495, 2019.
[29] L.Li, S.Fang, S.Bai, S.Xu, J.Cheng, Xi.Chen,” Effective Link Prediction Based on Community Relationship Strength,” IEEE Access, Volume:7, pp.43233 - 43248, 2019.
[30] Z.Liu, Y.Li, H.Liu,”Link Prediction in Evolving Networks Base on Information Propagation,”IEEE Access, Volume:7, pp.140451-140459, 2019.
[31] L.Liu, Y.Zhang, S.Fu, F.Zhong, Jun Hu ; Pu Zhang,” ABNE: An Attention-Based Network Embedding for User Alignment Across Social Networks,” IEEE Access, Volume:7, pp.23595 - 23605, 2019.
[32] J.Lin, L.Zhang, M.He, H.Zhang, G.Liu, Xiuyuan,” Multi-Path Relationship Preserved Social Network Embedding,” IEEE Access, Volume:7, pp.26507 - 26518, 2019.
[33] R. Michalski, S. Palus, and P. Kazienko, “Matching organizational structure and social network extracted from email communication,” in Lecture Notes in Business Information Processing, vol. 87. Springer Berlin Heidelberg, 2011, pp. 197–206.
[34] M.Mirza, S.Osindero, “Conditional Generative Adversarial Nets,” arXiv CS, 2014.
[35] N.Mohamed-Ahmed, L.Chen, Y.Wang, B.Li, Y.Li, W.Liu,” DeepEye: Link prediction in dynamic networks based on non-negative matrix factorization,” IEEE Access, Volume:1, pp.19 - 33, 2018.
[36] A. Papadimitriou, P. Symeonidis, and Y. Manolopoulos, “Fast and accurate link prediction in social networking systems,” Journal of Systems and Software, vol. 85, no. 9, pp. 2119–2132, 2012.
[37] X.Pan, G.Xu, B.Wang, T.Zhang,” A Novel Community Detection Algorithm Based on Local Similarity of Clustering Coefficient in Social Networks,” IEEE Access, Volume:7, pp.121586 – 121598, 2019.
[38] F.Qian, Y.Gao, S.Zhao, J.Tang, Y.Zhang,” Combining topological properties and strong ties for link prediction,” IEEE Access, Volume:22, pp.595 - 608, 2017.
[39] J.Ren, H.Tian, Y.Lin, S.Fan, G.Nie, H.Wu, Fan Zhang,” Incentivized Social-Aware Proactive Device Caching With User Preference Prediction,”IEEE Access, Volume:7, pp.136148 – 136160, 2019.
[40] M.Sadiq- Khan, A.Wahid-Abdul-Wahab, T.Herawan,” Virtual Community Detection Through the Association between Prime Nodes in Online Social Networks and Its Application to Ranking Algorithms,” IEEE Access, Volume:4, pp. 9614 - 9624, 2016.
[41] J.Tang, S.Chang, C.Aggarwal, H.Liu,” Negative Link Prediction in Social Media,” CU, Computer Science, arXiv:1412.2723,2014.
[42] H.Wang, W.Hu, Z.Qiu, B.Wu,” An Event Detection Method for Social Networks Based on Evolution Fluctuations of Nodes,” IEEE Access, Volume:6, pp.12351 - 12359, 2017.
[43] D.Wei, “Prediction of Stock Price Based on LSTM Neural Network,” IEEE 2020.
[44] B.Xu, Lu Li, Jiaying Liu, Liangtian Wan, Xiangjie Kong, Feng Xia,” Disappearing Link Prediction in Scientific Collaboration Networks,” IEEE Access, Volume:6, pp.69702 - 69712, 2020.
[45] C.Yang, X.Shi, L.Jie, J.Han, “I Know You’ll Be Back: Interpretable New User Clustering and Churn Prediction on a Mobile Social Application,” in the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 914-922, ACM, 2018.
[46] L.Yang, Z.Zhang, X.Cai, L.Guo,” Citation Recommendation as Edge Prediction in Heterogeneous Bibliographic Network: A Network Representation Approach,” IEEE Volume:7, pp.23232 - 23239, 2019.
[47] W.Yuan, K.He, D.Guan, G.Han,” Edge-Dual Graph Preserving Sign Prediction for Signed Social Networks,” IEEE Access, Volume:5, pp.128 - 135, 2017.
[48] Y.Yang, H.Guo, Ti.Tian, H.Li,” Link prediction in brain networks based on a hierarchical random graph model,” Tsinghua Science and Technology, Volume:20, pp.306 - 315, 2020.
[49] Yunjun Yu, Junfei Cao, Jianyong Zhu, “An LSTM Short-Term Solar Irradiance Forecasting Under Complicated Weather Conditions,” IEEE Acess, pp.145651 – 145666, 2019.
[50] H.Zhang, T.Xu, Ho.Li, S.Zhang, X.Wang, X.Huang, D.Metaxas, “StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks,” ICCV 2017.
[51] Z.Zhang, R.Sun, K.Raymond-Choo, K.Fan, W.Wu, M.Zhang, C.Zhao,”A Novel Social Situation Analytics-Based Recommendation Algorithm for Multimedia Social Networks,” IEEE Access, Volume:7, pp.117749 - 117760, 2019.
[52] J.Zhu, T.Park, P.Isola, A.A. Efros, “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks,” arXiv CS, 2018.
論文使用權限
  • 同意紙本無償授權給館內讀者為學術之目的重製使用,於2020-08-05公開。
  • 同意授權瀏覽/列印電子全文服務,於2020-08-05起公開。


  • 若您有任何疑問,請與我們聯絡!
    圖書館: 請來電 (02)2621-5656 轉 2486 或 來信