目前，社交網絡平台已逐漸形成人們傳遞信息不可或缺的渠道。 借助社交網絡平台，它們可以幫助用戶傳達自己的想法，並使社交平台上的文章內容多樣化。雖然可以加深民主思想，但也容易因為這樣的便利而產生仇恨。 由於信息量巨大，文章中總是充滿了用戶的不同觀點，這使得社會學家很難了解當前對某些事件的看法。 大多數人的想法都是支持或反對的，所以用戶的立場相對來說非常重要。本研究探討了台灣最具影響力的討論平台PTT論壇中不同意見群體之間的互動關係。 利用用戶反應和時間特徵的數據，我們識別有影響力的意見領袖並對群體進行分類。 我們通過單值分解 (SVD) 解決缺失數據的挑戰，並開發交互式可視化工具來觀察群體關係。 該研究增強了對社交媒體動態和輿論動態的理解。

At present, social network platforms have gradually formed an indispensable channel for people to transmit information. Thanks to social network platforms, they can help users to convey their own ideas, and make the content of articles on social platforms diverse. Although it can deepen democratic thinking, it is easy to generate hatred because of such convenience. Because of the huge amount of information, the articles are always full of different opinions of users, which makes it difficult for sociologists to know the current opinions on certain events. Most people think for or against, so the user’s position is relatively very important.This research explores the interaction relationships among different opinion groups within the PTT forum, Taiwan’s most influential discussion platform. Using data from users’ reactions and temporal characteristics, we identify influential opinion leaders and classify groups. We address missing data challenges through singular Value Decomposition (SVD) and develop an interactive visualization tool to observe group relationships. The study enhances understanding of social media dynamics and opinion dynamics.

[1] In Derek L. Hansen, Ben Shneiderman, and Marc A. Smith, editors, Analyzing Social Media Networks with NodeXL, pages 277–284. Morgan Kaufmann, Boston, 2011.

[2] Taushif Anwar, V Uma, and Gautam Srivastava. Rec-cfsvd++: Implementing recommendation system using collaborative filtering and singular value decomposition (svd)++. International Journal of Information Technology & Decision Making, 20(04):1075–1093, 2021.

[3] Ana Belen Barrag ´ ans-Mart ´ ´ınez, Enrique Costa-Montenegro, Juan C. Burguillo, Marta Rey-Lopez, Fernando A. Mikic-Fonte, and Ana Peleteiro. A hybrid content-based and ´ item-based collaborative filtering approach to recommend tv programs enhanced with singular value decomposition. Information Sciences, 180(22):4290–4311, 2010.

[4] Charalampos Chelmis and Viktor K. Prasanna. Social networking analysis: A state of the art and the effect of semantics. In 2011 IEEE Third International Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third International Conference on Social Computing, pages 531–536, 2011.

[5] Siming Chen, Shuai Chen, Lijing Lin, Xiaoru Yuan, Jie Liang, and Xiaolong Zhang. E-map: A visual analytics approach for exploring significant event evolutions in social media. In 2017 IEEE Conference on Visual Analytics Science and Technology (VAST), pages 36–47, 2017.

[6] Ju-Han Chuang and Shukai Hsieh. Stance classification on ptt comments. 2015.

[7] William Claster, Subana Shanmuganathan, and Philip Sallis. Wine tasting and a novel approach to cluster analysis. In 2010 Fourth Asia International Conference on Mathematical/Analytical Modelling and Computer Simulation, pages 152–157, 2010.

[8] Nicholas Diakopoulos, Mor Naaman, and Funda Kivran-Swaine. Diamonds in the rough: Social media visual analytics for journalistic inquiry. In 2010 IEEE Symposium on Visual Analytics Science and Technology, pages 115–122, 2010.

[9] Marian Dork, Daniel Gruen, Carey Williamson, and Sheelagh Carpendale. A visual ¨ backchannel for large-scale events. IEEE Transactions on Visualization and Computer Graphics, 16(6):1129–1138, 2010.

[10] A. Hernandez-Suarez, G. Sanchez-Perez, V. Martinez-Hernandez, H. Perez-Meana, K. Toscano-Medina, M. Nakano, and V. Sanchez. Predicting political mood tendencies based on twitter data. May 2017. Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017 ; Conference date: 26-05-2017.

[11] P. Howland and H. Park. Generalizing discriminant analysis using the generalized singular value decomposition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(8):995–1006, 2004.

[12] Sihem Jebari, Abir Smiti, and Aymen Louati. Af-dbscan: An unsupervised automatic fuzzy clustering method based on dbscan approach. In 2019 IEEE International Work Conference on Bioinspired Intelligence (IWOBI), pages 000001–000006, 2019.

[13] Sun Jing-Tao and Zhang Qiu-Yu. Completion of multiview missing data based on multi-manifold regularised non-negative matrix factorisation. Artificial Intelligence Review, 53(7):5411–5428, 2020.

[14] P Kanchanamala, V Vineela, and G Neelima. Traffic data analysis using hadoop based hierarchical clustering. In International Conference on Recent Trends in Engineering, Science Technology - (ICRTEST 2016), pages 1–4, 2016.

[15] Igor G. Khanykov. Technique for acceleration of classical ward’s method for clustering of image pixels. In 2019 International Russian Automation Conference (RusAutoCon), pages 1–6, 2019.

[16] Roma Kriauciˇ unien ¯ e, Jefferey Roux, and Migl ˙ e Lauci ˙ ut¯ e. Stance taking in social ˙ media: the analysis of the comments about us presidential candidates on facebook and twitter. Verbum, 9:21–30, 12 2018.

[17] Mimoun Lamrini and Mohamed Yassin Chkouri. Decomposition and visualization of high-dimensional data in a two dimensional interface. In 2019 1st International Conference on Smart Systems and Data Science (ICSSD), pages 1–5, 2019.

[18] Longzhuang Li and Douglas Boulware. High-order tensor decomposition for largescale data analysis. In 2015 IEEE International Congress on Big Data, pages 665–668, 2015.

[19] Qipeng Liu, Jiuhua Zhao, and Xiaofan Wang. A multi-agent model of opinion formation with group polarization. In Proceedings of the 33rd Chinese Control Conference, pages 1680–1685, 2014.

[20] Aibek Makazhanov and Davood Rafiei. Predicting political preference of twitter users. In 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2013), pages 298–305, 2013.

[21] Adam Marcus, Michael S. Bernstein, Osama Badar, David R Karger, Samuel Madden, and Rob Miller. Twitinfo: aggregating and visualizing microblogs for event exploration. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2011.

[22] Nisha and Puneet Jai Kaur. Cluster quality based performance evaluation of hierarchical clustering method. In 2015 1st International Conference on Next Generation Computing Technologies (NGCT), pages 649–653, 2015.

[23] Hongyi Peng and Siming Zhu. Handling of incomplete data sets using ica and som in data mining. Neural Computing and Applications, 16:167–172, 2007.

[24] A. Jaya Mabel Rani and Latha Parthipan. Clustering analysis by improved particle swarm optimization and k-means algorithm. In IET Chennai 3rd International on Sustainable Energy and Intelligent Systems (SEISCON 2012), pages 1–6, 2012.

[25] Keyvan Vahidy Rodpysh, Seyed Javad Mirabedini, and Touraj Banirostam. Employing singular value decomposition and similarity criteria for alleviating cold start and sparse data in context-aware recommender systems. Electronic Commerce Research, 23(2):681–707, 2023.

[26] J. Rosca, T. Gerkmann, and D.-C. Balcan. Statistical inference of missing speech data in the ica domain. In 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, volume 5, pages V–V, 2006.

[27] Kristen A. Severson, Mark C. Molaro, and Richard D. Braatz. Principal component analysis of process datasets with missing values. Processes, 5(3), 2017.

[28] David A. Shamma. Tweetgeist : Can the twitter timeline reveal the structure of broadcast events ? 2009.

[29] Zeinab Sharifi, Mansoor Rezghi, and Mahdi Nasiri. A new algorithm for solving data sparsity problem based-on non negative matrix factorization in recommender systems. In 2014 4th International Conference on Computer and Knowledge Engineering (ICCKE), pages 56–61. IEEE, 2014.

[30] Priyanka Sinha, Lipika Dey, Pabitra Mitra, and Dilys Thomas. A hierarchical clustering algorithm for characterizing social media users. In Companion Proceedings of the Web Conference 2020, pages 353–362, 2020.

[31] Kate Starbird, Leysia Palen, Amanda Lee Hughes, and Sarah Vieweg. Chatter on the red: what hazards threat reveals about the social life of microblogged information. In Conference on Computer Supported Cooperative Work, 2010.

[32] Nisha Tanwani, Sandesh Kumar, Akhtar Hussain Jalbani, Saima Soomro, Muhammad Ibrahim Channa, and Zeeshan Nizamani. Student opinion mining regarding educational system using facebook group. In 2017 First International Conference on Latest trends in Electrical Engineering and Computing Technologies (INTELLECT), pages 1–5, 2017.

[33] Joost R van Ginkel. Handling missing data in principal component analysis using multiple imputation. In Essays on Contemporary Psychometrics, pages 141–161. Springer, 2023.

[34] Sarah Vieweg, Amanda Lee Hughes, Kate Starbird, and Leysia Palen. Microblogging during two natural hazards events: what twitter may contribute to situational awareness. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2010.

[35] Shujun Wang and Xiaoshen Cai. Literature analysis of xi jinping’s important expositions on ideological and political education based on cluster analysis. In 2021 2nd International Conference on Education, Knowledge and Information Management (ICEKIM), pages 792–796, 2021.

[36] Yingcai Wu, Shixia Liu, Kai Yan, Mengchen Liu, and Fangzhao Wu. Opinionflow: Visual analysis of opinion diffusion on social media. IEEE Transactions on Visualization and Computer Graphics, 20(12):1763–1772, 2014.

[37] Qing Yang, Ye Liu, Dongxu Zhang, and Chang Liu. Improved k-means algorithm to quickly locate optimum initial clustering number k. In Proceedings of the 30th Chinese Control Conference, pages 3319–3322, 2011.

[38] Chunsheng Ye and Xuan Zhao. Automated operational modal analysis based on dbscan clustering. In 2020 International Conference on Intelligent Transportation, Big Data Smart City (ICITBS), pages 864–869, 2020.

[39] Jun Yin, Jianye Zhang, Degao Li, Tianjun Wang, and Kang Jing. Big data cleaning model of smart grid based on tensor tucker decomposition. In 2020 International Conference on Big Data Artificial Intelligence Software Engineering (ICBASE), pages 166–169, Oct 2020.

[40] Zhiwen Yu, Hantao Chen, Jane You, Jiming Liu, Hau-San Wong, Guoqiang Han, and Le Li. Adaptive fuzzy consensus clustering framework for clustering analysis of cancer data. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 12(4):887–901, 2015.

[41] Feng Zhou, Fernando De la Torre, and Jessica K. Hodgins. Hierarchical aligned cluster analysis for temporal clustering of human motion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(3):582–596, 2013.