進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-2507201816301900
論文名稱(中文) 社群網路最具影響力小編探勘及其貼文時間規劃
論文名稱(英文) Mining Influential Who-to-Post and When-to-Post Curators on Social Networks
校院名稱 成功大學
系所名稱(中) 統計學系
系所名稱(英) Department of Statistics
學年度 106
學期 2
出版年 107
研究生(中文) 夏婕禎
研究生(英文) Chieh-Cheng Hsia
學號 R26051066
學位類別 碩士
語文別 中文
論文頁數 74頁
口試委員 指導教授-李政德
口試委員-黃仁暐
口試委員-解巽評
中文關鍵字 社群網路分析  特徵工程  影響力最大化 
英文關鍵字 Social Network Analysis  Feature Engineering  Influence Maximization 
學科別分類
中文摘要 線上社群網路的發展和成熟,使得社群平台已成為有效宣傳且行銷商品及品牌形象之重要媒介,而社群小編 (Curator) 也因此需求而成為各商家進行網路行銷所不可或缺的代言人。如何尋找最具未來影響力的小編,並為其規劃貼文時間,使得具未來影響力的小編們能在未來時間內於其最佳貼文時段發佈貼文,以獲取未來時間之最多回覆數?本研究將針對此二主軸進行四個不同面向的探討與分析。針對預測最具未來影響力之社群小編,我們進行貼文者未來影響力排序預測、以及未來意見領袖預測;針對最具未來影響力之貼文時間規劃,我們將其分為累積影響力與限時影響力來進行之最佳貼文時段預測。本研究提出一系列與小編影響力相關之基本與進階行為特徵,並透過社群網路嵌入特徵向量學習方法 (network embedding) 進一步學習用戶間彼此影響關係的到學習特徵。在預測方法上,我們採用學習排序演算法,並提出自我訓練 (self-training) 與交互訓練 (mutual-training) 來解決影響力類別極度不平衡之問題。我們基於大規模 Facebook 用戶與貼文之互動資料來進行實驗評估,結果顯示我們所提方法在四個不同面向的研究問題上,相較過去預測方法及普通特徵擷取,皆可獲得明顯準確率之提升;其中預測最具未來影響力小編之 F1 可高達 0.875, 此外,我們所規劃預測之最佳貼文時段可使小編的限時影響力提升將近 3 倍。
英文摘要 Curators on the social networking sites become more and more prominent and indispensable nowadays. Gradually, they come to be the voice in the business’s online marketing field. The problem that how to find the most future-influential curators and plan the best posting time for them, notwithstanding, has been hidden and under-explored as yet. In this study, we initiate to analyze this problem with those two primary concerns from four distinct dimensions. To find the most future-influential curators, we consider this subproblem from following two dimensions, Future Influence Ranking Prediction (FIRP) and Future Influential Leader Prediction (FILP). To plan the best posting time for the curators, similarly, we consider this part with two dimensions below, Accumulated Influence Post-time Scheduling and Limited Influence Post-time Scheduling. We aim at predicting the future influential curator with a series of basic and advanced self-defined features. Based on network embedding, we add learned features to capture the connection between users. To deal with the problem, we implement Learning to Rank algorithm and two newly devised ones, self-training algorithm as well as mutual-training algorithm, which are served to become the solution for imbalanced data. With the experiments on large-scale Facebook data for evaluation, we find that the proposed methods significantly outperform conventional prediction settings. The F1 score in predicting the most future influential curators can be up to 0.875. Also, in the part of planning the best posting time, the result shows in comparison with the overall performance of curators, the limited influence of the curators in our planned time can be boosted up to three times.
論文目次 摘要 i
英文延伸摘要 ii
誌謝 vii
目錄 viii
表目錄 x
圖目錄 xi

第1章.緒論 1
1.1.背景...................................... 1
1.2.動機...................................... 2
1.3.研究問題................................ 2
1.4.研究挑戰................................ 3
1.5.方法概述.............................. 4
1.6.潛在應用................................ 5

第 2 章.相關研究 7
2.1.影響力最大化................................ 7
2.2.影響力量化.................................. 8
2.3.影響力預測.................................. 9
2.4.貼文時間規劃................................ 11

第 3 章. 研究問題敘述 13
3.1.符號定義................................... 13
3.2. 預測最具未來影響力之小編 ........................ 14
3.2.1. Task1:社群未來影響力排序預測................. 15
3.2.2. Task2:社群未來意見領袖預測................... 17
3.3. 最具影響力之貼文時段預測........................ 19
3.3.1. Task3:累積影響力之貼文時段預測................ 19
3.3.2. Task4:限時影響力之貼文時段預測................ 20

第 4 章.研究方法 23
4.1. 研究架構與方法流程............................ 23
4.2.特徵工程................................... 24
4.2.1. 基本特徵擷取............................ 24
4.2.2. 進階特徵擷取............................ 25
4.2.3. 特徵表示學習............................ 27
4.3. Task1:社群未來影響力排序預測...................... 32
4.3.1.特徵向量............................... 33
4.3.2.預測模型............................... 34
4.4. Task2:社群未來意見領袖預測....................... 34
4.4.1. 訓練資料演算法-自我訓練演算法................ 36
4.4.2. 訓練資料演算法-交互訓練演算法................ 38
4.5. Task3&4:最具影響力之貼文時段預測.................. 40
4.5.1. 不限時段貼文............................ 40
4.5.2. 相同時段貼文............................ 44

第 5 章.實驗評估 47
5.1.實驗目的................................... 47
5.2. 資料集與摘要統計量............................ 47
5.3. Task1:社群未來影響力排序預測...................... 49
5.3.1.實驗設定............................... 49
5.3.2.評估指標............................... 50
5.3.3.實驗結果............................... 50
5.4. Task2:社群未來意見領袖預測....................... 57
5.4.1.實驗設定............................... 57
5.4.2.評估指標............................... 57
5.4.3.實驗結果............................... 58
5.5. Task3&4:最具影響力之貼文時段預測 .................. 63
5.5.1.  實驗設定............................... 63
5.5.2.  評估指標............................... 64
5.5.3.  實驗結果............................... 66

第6章.結論 70

參考文獻 71
參考文獻 參考文獻

[1] Nitin Agarwal, Huan Liu, Lei Tang, and Philip S. Yu. Identifying the influential blog- gers in a community. In Proceedings of the 2008 International Conference on Web Search and Data Mining, WSDM ’08, pages 207–218, 2008.
[2] Mohamed Ahmed, Stella Spagna, Felipe Huici, and Saverio Niccolini. A peek into the future: Predicting the evolution of popularity in user generated content. In Proceedings of the Sixth ACM International Conference on Web Search and Data Mining, WSDM ’13, pages 607–616, 2013.
[3] Eytan Bakshy, Jake M. Hofman, Winter A. Mason, and Duncan J. Watts. Everyone’s an influencer: Quantifying influence on twitter. In Proceedings of the Fourth ACM International Conference on Web Search and Data Mining, WSDM ’11, pages 65–74, 2011.
[4] Michael S. Bernstein, Eytan Bakshy, Moira Burke, and Brian Karrer. Quantifying the invisible audience in social networks. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI ’13, pages 21–30, 2013.
[5] Smriti Bhagat, Amit Goyal, and Laks V.S. Lakshmanan. Maximizing product adoption in social networks. In Proceedings of the Fifth ACM International Conference on Web Search and Data Mining, WSDM ’12, pages 603–612, 2012.
[6] Christopher J.C. Burges. From RankNet to LambdaRank to LambdaMART: An Overview. Microsoft Research Technical Report MSR-TR-2010-82, 2010.
[7] Meeyoung Cha, Alan Mislove, and Krishna P. Gummadi. A measurement-driven anal- ysis of information propagation in the flickr social network. In Proceedings of the 18th International Conference on World Wide Web, WWW ’09, pages 721–730, 2009.
[8] Wei Chen, Yajun Wang, and Siyu Yang. Efficient influence maximization in social net- works. In Proceedings of the 15th ACM SIGKDD International Conference on Knowl- edge Discovery and Data Mining, KDD ’09, pages 199–208, 2009.
[9] Justin Cheng, Lada Adamic, P. Alex Dow, Jon Michael Kleinberg, and Jure Leskovec. Can cascades be predicted? In Proceedings of the 23rd International Conference on World Wide Web, WWW ’14, pages 925–936, 2014.
[10] Peng Cui, Fei Wang, Shaowei Liu, Mingdong Ou, Shiqiang Yang, and Lifeng Sun. Who should share what?: Item-level social influence prediction for users and posts ranking. In Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’11, pages 185–194, 2011.
[11] Shuai Gao, Jun Ma, and Zhumin Chen. Modeling and predicting retweeting dynamics on microblogging platforms. In Proceedings of the Eighth ACM International Confer- ence on Web Search and Data Mining, WSDM ’15, pages 107–116, 2015.
[12] Rumi Ghosh and Kristina Lerman. Predicting influential users in online social networks. CoRR, abs/1005.4882, 2010.
[13] Amit Goyal, Francesco Bonchi, and Laks V. S. Lakshmanan. A data-based approach to social influence maximization. Proc. VLDB Endow., 5(1):73–84, September 2011.
[14] Amit Goyal, Francesco Bonchi, and Laks V.S. Lakshmanan. Discovering leaders from community actions. In Proceedings of the 17th ACM Conference on Information and Knowledge Management, CIKM ’08, pages 499–508, 2008.
[15] Aditya Grover and Jure Leskovec. node2vec: Scalable feature learning for networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016.
[16] C. T. Ho, C. T. Li, and S. D. Lin. Modeling and visualizing information propagation in a micro-blogging platform. In 2011 International Conference on Advances in Social Networks Analysis and Mining, pages 328–335, July 2011.
[17] Tomoharu Iwata, Amar Shah, and Zoubin Ghahramani. Discovering latent influence in online social activities via shared cascade poisson processes. In Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’13, pages 266–274, 2013.
[18] Mohammad Reza Karimi, Erfan Tavakoli, Mehrdad Farajtabar, Le Song, and Manuel Gomez Rodriguez. Smart broadcasting: Do you want to be seen? In Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, pages 1635–1644, 2016.
[19] David Kempe, Jon Kleinberg, and Éva Tardos. Maximizing the spread of influence through a social network. In Proceedings of the Ninth ACM SIGKDD International Con- ference on Knowledge Discovery and Data Mining, KDD ’03, pages 137–146, 2003.
[20] Siddharth Krishnan, Patrick Butler, Ravi Tandon, Jure Leskovec, and Naren Ramakr- ishnan. Seeing the forest for the trees: New approaches to forecasting cascades. In Proceedings of the 8th ACM Conference on Web Science, WebSci ’16, pages 249–258, 2016.
[21] Andrey Kupavskii, Liudmila Ostroumova, Alexey Umnov, Svyatoslav Usachev, Pavel Serdyukov, Gleb Gusev, and Andrey Kustarev. Prediction of retweet cascade size over time. In Proceedings of the 21st ACM International Conference on Information and Knowledge Management, CIKM ’12, pages 2335–2338, 2012.
[22] Theodoros Lappas, Evimaria Terzi, Dimitrios Gunopulos, and Heikki Mannila. Finding effectors in social networks. In Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’10, pages 1059–1068, 2010.
[23] Jure Leskovec, Andreas Krause, Carlos Guestrin, Christos Faloutsos, Jeanne Van- Briesen, and Natalie Glance. Cost-effective outbreak detection in networks. In Pro- ceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’07, pages 420–429, 2007.
[24] Xujun Li, Yezheng Liu, Yuanchun Jiang, and Xiao Liu. Identifying social influence in complex networks. Neurocomput., 210(C):141–154, October 2016.
[25] Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. Efficient estimation of word representations in vector space. CoRR, abs/1301.3781, 2013.
[26] Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg Corrado, and Jeffrey Dean. Distributed representations of words and phrases and their compositionality. In Proceedings of the 26th International Conference on Neural Information Processing Systems - Volume 2, NIPS’13, pages 3111–3119, 2013.
[27] Nemanja Spasojevic, Zhisheng Li, Adithya Rao, and Prantik Bhattacharyya. When- to-post on social networks. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’15, pages 2127–2136, 2015.
[28] Karthik Subbian, B. Aditya Prakash, and Lada Adamic. Detecting large reshare cas- cades in social networks. In Proceedings of the 26th International Conference on World Wide Web, WWW ’17, pages 597–605, 2017.
[29] Jie Tang, Jimeng Sun, Chi Wang, and Zi Yang. Social influence analysis in large- scale networks. In Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’09, pages 807–816, 2009.
[30] Youze Tang, Yanchen Shi, and Xiaokui Xiao. Influence maximization in near-linear time: A martingale approach. In Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data, SIGMOD ’15, pages 1539–1554, 2015.
[31] Youze Tang, Xiaokui Xiao, and Yanchen Shi. Influence maximization: Near-optimal time complexity meets practical efficiency. In Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data, SIGMOD ’14, pages 75–86, 2014.
[32] Ali Zarezade, Utkarsh Upadhyay, Hamid R. Rabiee, and Manuel Gomez-Rodriguez. Redqueen: An online algorithm for smart broadcasting in social networks. In Pro- ceedings of the Tenth ACM International Conference on Web Search and Data Mining, WSDM ’17, pages 51–60, 2017.
[33] Chao Zhang, Lidan Shou, Ke Chen, Gang Chen, and Yijun Bei. Evaluating geo-social influence in location-based social networks. In Proceedings of the 21st ACM Inter- national Conference on Information and Knowledge Management, CIKM ’12, pages 1442–1451, 2012.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2023-07-01起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2023-07-01起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw