學術產出-Theses
Article View/Open
Publication Export
-
題名 應用Personalized PageRank與RFM模式區隔比特幣投資者類群
Classifying the segmentation of Bitcoin investors via personalized PageRank and RFM model作者 林聖翔
Lin, Sheng-Hsiang貢獻者 楊建民<br>洪為璽
Yang, Chien-Min<br>Hung, Wei-Hsi
林聖翔
Lin, Sheng-Hsiang關鍵詞 比特幣
投資者區隔
Personalized PageRank
RFM模式
K-Means
Bitcoin
Investor segmentation
Personalized PageRank
RFM model
K-Means日期 2019 上傳時間 7-Aug-2019 16:07:03 (UTC+8) 摘要 比特幣是一種對等式架構(peer-to-peer, p2p)之去中心化的貨幣系統，為目前區塊鏈技術最為人所知的應用，近年來隨著網路及大眾媒體的傳播，比特幣受到投資人的青睞，因此交易活動逐年劇增，匯率也水漲船高，而累積的紀錄構成了龐大的交易網絡，因此本研究欲透過比特幣區塊鏈的交易資料來了解比特幣投資市場究竟由哪些類型的投資者所組成，市場又被哪些投資者主宰。本研究將使用Personalized PageRank演算法來評估比特幣投資者於交易網絡中的重要性，意即指節點在整體網絡中的交易量、連結數等扮演的角色及份量。我們將以RFM (Recency, Frequency, Monetary) 模型計算得出投資者的初始節點評分，並透過網絡架構令投資者的評分因交易連結傳遞。而為區隔不同類型的投資者，我們利用K-Means分群演算法以三個維度：投資者的網絡重要性、投資者擔任交易輸入方與輸出方的兩種交易模式之RFM評分，對比特幣的投資者進行分群。本研究以區塊高度自第514,988區塊至第521,639區塊共6,652個區塊的8,383,945筆交易資料建構比特幣投資者的交易網絡，將投資者分為5個群集：活躍投資者、穩定投資者、消極投資者、潛在出場者、新進投資者，其中活躍投資者主宰了整個比特幣市場，該群集的人數為整個市場的27.4%，並作為交易網絡中的樞紐，貢獻了整個網絡60% 的重要性，更貢獻了整個市場2/3的交易量及85%的交易金額。透過本研究提出之方法，能區隔比特幣或區塊鏈相關應用上的投資者，且本研究歸納之投資者類型，將能更了解比特幣交易市場中不同類型的投資者組成。
Bitcoin is a decentralized peer-to-peer (p2p) currency system, which is the most well-known application of blockchain technology. In recent years, with the spread of the Internet and mass media, Bitcoin is favored by investors. As the result, the trading activity of Bitcoin has increased dramatically year by year, the exchange rate has also risen. Because of the accumulated record constitutes a huge trading network, we want to understand Bitcoin market through the transaction information of Bitcoin’s blockchain. We want to explore what kinds of investors construct Bitcoin investment market and which investors cluster dominates the market.This study uses the Personalized PageRank algorithm to assess the importance of Bitcoin investors in the trading network, which means the weight of the node`s trading volume and number of links in the overall network. We will use the RFM (Recency, Frequency, Monetary) model to calculate the initial score of the investor, the investor`s score will be transmitted through the transaction link among the network structure. In order to distinguish different types of investors, the K-Means clustering algorithm is used with three dimensions: the investor`s network importance, the investor`s RFM score of different transaction modes (input/output).In this study, we constructed a trading network graph of Bitcoin investors by blocks height 514,988 to 521,639 which have 6,652 blocks and 8,383,945 transactions. We divided investors into five clusters: active investors, stable investors, passive investors, potential abandoner, new investors. Active investors cluster dominates the market by 27.5% user amount, 60% network importance, 2/3 trading volume and 85% transaction amount of whole market. Through the methods proposed in this study, investors can be distinguished in Bitcoin or blockchain-related applications. Moreover, the investors types summarized in this study will be able to better understand the different properties of each investor cluster in the Bitcoin trading market.參考文獻 中文文獻方贊宗. (2017). 由區塊鏈資料探討比特幣特性. 臺灣大學資訊管理學研究所學位論文, 1-63.英文文獻Anderberg, M. R. (2014). Cluster analysis for applications: probability and mathematical statistics: a series of monographs and textbooks (Vol. 19). Academic press.Androulaki, E., Karame, G. O., Roeschlin, M., Scherer, T., & Capkun, S. (2013, April). Evaluating user privacy in bitcoin. In International Conference on Financial Cryptography and Data Security (pp. 34-51)Baur, D. G., Hong, K., & Lee, A. D. (2018). Bitcoin: Medium of exchange or speculative assets?. Journal of International Financial Markets, Institutions and Money, 54, 177-189.Borgatti, S. P., & Everett, M. G. (2006). A graph-theoretic perspective on centrality. Social networks, 28(4), 466-484.Cheng, C. H., & Chen, Y. S. (2009). Classifying the segmentation of customer value via RFM model and RS theory. Expert systems with applications, 36(3), 4176-4184.Christin, N. (2013, May). Traveling the Silk Road: A measurement analysis of a large anonymous online marketplace. In Proceedings of the 22nd international conference on World Wide Web (pp. 213-224). ACM.Dwyer, F. R. (1997). Customer lifetime valuation to support marketing decision making. Journal of Direct Marketing, 11(4), 6-13.Fleder, M., Kester, M. S., & Pillai, S. (2015). Bitcoin transaction graph analysis. arXiv preprint arXiv:1502.01657.Gerlach, J. C., Demos, G., & Sornette, D. (2018). Dissection of Bitcoin`s Multiscale Bubble History from January 2012 to February 2018. arXiv preprint arXiv:1804.06261.Glaser, F., Zimmermann, K., Haferkorn, M., Weber, M., & Siering, M. (2014). Bitcoin-asset or currency? revealing users` hidden intentions.Gyöngyi, Z., Garcia-Molina, H., & Pedersen, J. (2004, August). Combating web spam with trustrank. In Proceedings of the Thirtieth international conference on Very large data bases-Volume 30 (pp. 576-587). VLDB Endowment.Harrigan, M., & Fretter, C. (2016, July). The unreasonable effectiveness of address clustering. In Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld), 2016 Intl IEEE Conferences (pp. 368-373).Hughes, A. M. (1996). Boosting response with RFM. Marketing Tools, 4-8.Kodinariya, T. M., & Makwana, P. R. (2013). Review on determining number of Cluster in K-Means Clustering. International Journal, 1(6), 90-95.Kondor, D., Pósfai, M., Csabai, I., & Vattay, G. (2014). Do the rich get richer? An empirical analysis of the Bitcoin transaction network. PloS one, 9(2), e86197.Krishnan, V., & Raj, R. (2006, August). Web spam detection with anti-trust rank. In AIRWeb (Vol. 6, pp. 37-40).Meiklejohn, S., Pomarole, M., Jordan, G., Levchenko, K., McCoyy, D., Voelker, G. M., & Savage, S. (2013). A Fistful of Bitcoins: Characterizing Payments Among Men with No Names. Proceedings of the 2013 ACM Conference on Internet Measurement Conference. (pp. 127-140). Barcelona, Spain.Miglautsch, J. R. (2000). Thoughts on RFM scoring. Journal of Database Marketing & Customer Strategy Management, 8(1), 67-72.Mohurle, S., & Patil, M. (2017). A brief study of wannacry threat: Ransomware attack 2017. International Journal of Advanced Research in Computer Science, 8(5).Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system.Nick, J. D. (2015). Data-driven de-anonymization in bitcoin (Master`s thesis, ETH-Zürich)Page, L., Brin, S., Motwani, R., & Winograd, T. (1999). The PageRank citation ranking: Bringing order to the web. Stanford InfoLab.Reid, F., & Harrigan, M. (2011, October). An analysis of anonymity in the bitcoin system. In Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), 2011 IEEE Third International Conference on (pp. 1318-1326).Shih, Y. Y., & Liu, C. Y. (2003). A method for customer lifetime value ranking—Combining the analytic hierarchy process and clustering analysis. Journal of Database Marketing & Customer Strategy Management, 11(2), 159-172.Sohrabi, B., & Khanlari, A. (2007). Customer lifetime value (CLV) measurement based on RFM model.Wei, J. T., Lin, S. Y. and Wu, H. H. (2010), “A review of the application of RFMmodel,” African Journal of Business Management, 4(19), 4199-4206.Xue, T., Yuan, Y., & Wang, C. (2018, June). An Approach for Evaluating User Participation in Bitcoin. In 2018 IEEE Third International Conference on Data Science in Cyberspace (DSC) (pp. 858-864).Yelowitz, A., & Wilson, M. (2015). Characteristics of Bitcoin users: an analysis of Google search data. Applied Economics Letters, 22(13), 1030-1036.網際網路S. Lui, 2013. “The demographics of Bitcoin,” Simulacrum, at http://bit.ly/1FUXFru. 描述 碩士
國立政治大學
資訊管理學系
106356022資料來源 http://thesis.lib.nccu.edu.tw/record/#G0106356022 資料類型 thesis dc.contributor.advisor 楊建民<br>洪為璽 zh_TW dc.contributor.advisor Yang, Chien-Min<br>Hung, Wei-Hsi en_US dc.contributor.author (Authors) 林聖翔 zh_TW dc.contributor.author (Authors) Lin, Sheng-Hsiang en_US dc.creator (作者) 林聖翔 zh_TW dc.creator (作者) Lin, Sheng-Hsiang en_US dc.date (日期) 2019 en_US dc.date.accessioned 7-Aug-2019 16:07:03 (UTC+8) - dc.date.available 7-Aug-2019 16:07:03 (UTC+8) - dc.date.issued (上傳時間) 7-Aug-2019 16:07:03 (UTC+8) - dc.identifier (Other Identifiers) G0106356022 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/124711 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 資訊管理學系 zh_TW dc.description (描述) 106356022 zh_TW dc.description.abstract (摘要) 比特幣是一種對等式架構(peer-to-peer, p2p)之去中心化的貨幣系統，為目前區塊鏈技術最為人所知的應用，近年來隨著網路及大眾媒體的傳播，比特幣受到投資人的青睞，因此交易活動逐年劇增，匯率也水漲船高，而累積的紀錄構成了龐大的交易網絡，因此本研究欲透過比特幣區塊鏈的交易資料來了解比特幣投資市場究竟由哪些類型的投資者所組成，市場又被哪些投資者主宰。本研究將使用Personalized PageRank演算法來評估比特幣投資者於交易網絡中的重要性，意即指節點在整體網絡中的交易量、連結數等扮演的角色及份量。我們將以RFM (Recency, Frequency, Monetary) 模型計算得出投資者的初始節點評分，並透過網絡架構令投資者的評分因交易連結傳遞。而為區隔不同類型的投資者，我們利用K-Means分群演算法以三個維度：投資者的網絡重要性、投資者擔任交易輸入方與輸出方的兩種交易模式之RFM評分，對比特幣的投資者進行分群。本研究以區塊高度自第514,988區塊至第521,639區塊共6,652個區塊的8,383,945筆交易資料建構比特幣投資者的交易網絡，將投資者分為5個群集：活躍投資者、穩定投資者、消極投資者、潛在出場者、新進投資者，其中活躍投資者主宰了整個比特幣市場，該群集的人數為整個市場的27.4%，並作為交易網絡中的樞紐，貢獻了整個網絡60% 的重要性，更貢獻了整個市場2/3的交易量及85%的交易金額。透過本研究提出之方法，能區隔比特幣或區塊鏈相關應用上的投資者，且本研究歸納之投資者類型，將能更了解比特幣交易市場中不同類型的投資者組成。 zh_TW dc.description.abstract (摘要) Bitcoin is a decentralized peer-to-peer (p2p) currency system, which is the most well-known application of blockchain technology. In recent years, with the spread of the Internet and mass media, Bitcoin is favored by investors. As the result, the trading activity of Bitcoin has increased dramatically year by year, the exchange rate has also risen. Because of the accumulated record constitutes a huge trading network, we want to understand Bitcoin market through the transaction information of Bitcoin’s blockchain. We want to explore what kinds of investors construct Bitcoin investment market and which investors cluster dominates the market.This study uses the Personalized PageRank algorithm to assess the importance of Bitcoin investors in the trading network, which means the weight of the node`s trading volume and number of links in the overall network. We will use the RFM (Recency, Frequency, Monetary) model to calculate the initial score of the investor, the investor`s score will be transmitted through the transaction link among the network structure. In order to distinguish different types of investors, the K-Means clustering algorithm is used with three dimensions: the investor`s network importance, the investor`s RFM score of different transaction modes (input/output).In this study, we constructed a trading network graph of Bitcoin investors by blocks height 514,988 to 521,639 which have 6,652 blocks and 8,383,945 transactions. We divided investors into five clusters: active investors, stable investors, passive investors, potential abandoner, new investors. Active investors cluster dominates the market by 27.5% user amount, 60% network importance, 2/3 trading volume and 85% transaction amount of whole market. Through the methods proposed in this study, investors can be distinguished in Bitcoin or blockchain-related applications. Moreover, the investors types summarized in this study will be able to better understand the different properties of each investor cluster in the Bitcoin trading market. en_US dc.description.tableofcontents 摘要 IAbstract II目錄 IV表次 V圖次 VI第一章 緒論 1第一節 研究背景與動機 1第二節 研究目的 4第二章 文獻探討 5第一節 比特幣的媒介性質及使用者特性 5第二節 RFM模型與顧客區隔 7第三節 網絡分析與PageRank 10第三章 研究方法 13第一節 資料蒐集與前處理 14第二節 建構投資者交易網絡圖與轉移矩陣 19第三節 評估投資者交易模式及網絡重要性 23第四節 比特幣投資者區隔 28第四章 研究結果 30第一節 投資者交易網絡 30第二節 投資者RFM交易模式 32第三節 投資者網絡重要性 35第四節 比特幣投資者區隔 40第五章 結論與未來發展方向 52第一節 結論與建議 52第二節 研究限制及未來展望 54參考文獻 55 zh_TW dc.format.extent 3366699 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0106356022 en_US dc.subject (關鍵詞) 比特幣 zh_TW dc.subject (關鍵詞) 投資者區隔 zh_TW dc.subject (關鍵詞) Personalized PageRank zh_TW dc.subject (關鍵詞) RFM模式 zh_TW dc.subject (關鍵詞) K-Means zh_TW dc.subject (關鍵詞) Bitcoin en_US dc.subject (關鍵詞) Investor segmentation en_US dc.subject (關鍵詞) Personalized PageRank en_US dc.subject (關鍵詞) RFM model en_US dc.subject (關鍵詞) K-Means en_US dc.title (題名) 應用Personalized PageRank與RFM模式區隔比特幣投資者類群 zh_TW dc.title (題名) Classifying the segmentation of Bitcoin investors via personalized PageRank and RFM model en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) 中文文獻方贊宗. (2017). 由區塊鏈資料探討比特幣特性. 臺灣大學資訊管理學研究所學位論文, 1-63.英文文獻Anderberg, M. R. (2014). Cluster analysis for applications: probability and mathematical statistics: a series of monographs and textbooks (Vol. 19). Academic press.Androulaki, E., Karame, G. O., Roeschlin, M., Scherer, T., & Capkun, S. (2013, April). Evaluating user privacy in bitcoin. In International Conference on Financial Cryptography and Data Security (pp. 34-51)Baur, D. G., Hong, K., & Lee, A. D. (2018). Bitcoin: Medium of exchange or speculative assets?. Journal of International Financial Markets, Institutions and Money, 54, 177-189.Borgatti, S. P., & Everett, M. G. (2006). A graph-theoretic perspective on centrality. Social networks, 28(4), 466-484.Cheng, C. H., & Chen, Y. S. (2009). Classifying the segmentation of customer value via RFM model and RS theory. Expert systems with applications, 36(3), 4176-4184.Christin, N. (2013, May). Traveling the Silk Road: A measurement analysis of a large anonymous online marketplace. In Proceedings of the 22nd international conference on World Wide Web (pp. 213-224). ACM.Dwyer, F. R. (1997). Customer lifetime valuation to support marketing decision making. Journal of Direct Marketing, 11(4), 6-13.Fleder, M., Kester, M. S., & Pillai, S. (2015). Bitcoin transaction graph analysis. arXiv preprint arXiv:1502.01657.Gerlach, J. C., Demos, G., & Sornette, D. (2018). Dissection of Bitcoin`s Multiscale Bubble History from January 2012 to February 2018. arXiv preprint arXiv:1804.06261.Glaser, F., Zimmermann, K., Haferkorn, M., Weber, M., & Siering, M. (2014). Bitcoin-asset or currency? revealing users` hidden intentions.Gyöngyi, Z., Garcia-Molina, H., & Pedersen, J. (2004, August). Combating web spam with trustrank. In Proceedings of the Thirtieth international conference on Very large data bases-Volume 30 (pp. 576-587). VLDB Endowment.Harrigan, M., & Fretter, C. (2016, July). The unreasonable effectiveness of address clustering. In Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld), 2016 Intl IEEE Conferences (pp. 368-373).Hughes, A. M. (1996). Boosting response with RFM. Marketing Tools, 4-8.Kodinariya, T. M., & Makwana, P. R. (2013). Review on determining number of Cluster in K-Means Clustering. International Journal, 1(6), 90-95.Kondor, D., Pósfai, M., Csabai, I., & Vattay, G. (2014). Do the rich get richer? An empirical analysis of the Bitcoin transaction network. PloS one, 9(2), e86197.Krishnan, V., & Raj, R. (2006, August). Web spam detection with anti-trust rank. In AIRWeb (Vol. 6, pp. 37-40).Meiklejohn, S., Pomarole, M., Jordan, G., Levchenko, K., McCoyy, D., Voelker, G. M., & Savage, S. (2013). A Fistful of Bitcoins: Characterizing Payments Among Men with No Names. Proceedings of the 2013 ACM Conference on Internet Measurement Conference. (pp. 127-140). Barcelona, Spain.Miglautsch, J. R. (2000). Thoughts on RFM scoring. Journal of Database Marketing & Customer Strategy Management, 8(1), 67-72.Mohurle, S., & Patil, M. (2017). A brief study of wannacry threat: Ransomware attack 2017. International Journal of Advanced Research in Computer Science, 8(5).Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system.Nick, J. D. (2015). Data-driven de-anonymization in bitcoin (Master`s thesis, ETH-Zürich)Page, L., Brin, S., Motwani, R., & Winograd, T. (1999). The PageRank citation ranking: Bringing order to the web. Stanford InfoLab.Reid, F., & Harrigan, M. (2011, October). An analysis of anonymity in the bitcoin system. In Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), 2011 IEEE Third International Conference on (pp. 1318-1326).Shih, Y. Y., & Liu, C. Y. (2003). A method for customer lifetime value ranking—Combining the analytic hierarchy process and clustering analysis. Journal of Database Marketing & Customer Strategy Management, 11(2), 159-172.Sohrabi, B., & Khanlari, A. (2007). Customer lifetime value (CLV) measurement based on RFM model.Wei, J. T., Lin, S. Y. and Wu, H. H. (2010), “A review of the application of RFMmodel,” African Journal of Business Management, 4(19), 4199-4206.Xue, T., Yuan, Y., & Wang, C. (2018, June). An Approach for Evaluating User Participation in Bitcoin. In 2018 IEEE Third International Conference on Data Science in Cyberspace (DSC) (pp. 858-864).Yelowitz, A., & Wilson, M. (2015). Characteristics of Bitcoin users: an analysis of Google search data. Applied Economics Letters, 22(13), 1030-1036.網際網路S. Lui, 2013. “The demographics of Bitcoin,” Simulacrum, at http://bit.ly/1FUXFru. zh_TW dc.identifier.doi (DOI) 10.6814/NCCU201900434 en_US
