以視覺化系統探索社群媒體中的使用者行為

A Visual Analytics System for Exploring User Behavior in Social Media

李柏彥(Li Po-Yen)

指導教授 ： 紀明德

國立政治大學/理學院/資訊科學系/碩士(2021年)

https://doi.org/10.6814/NCCU202100352

摘要

社群媒體是現代社會中獲取資訊的主要管道之一，同時也是人們發表意見的地方，然而，人們的意見經常被社群媒體上的意見領袖所左右，而意見領袖的聲量通常透過發表文章及評論來累積，但也有可能是特定集團抬轎而來，因此，本研究提出一個視覺化分析系統，將使用者之間的行為相似度表示成距離矩陣並排序，再以基於熱度圖的視覺化呈現，並且利用使用者與文章的二分關係將其繪製成具關聯矩陣風格的視覺化，結合上述兩者視覺化來觀察使用者活動情形並比較不同使用者類型的差異，可以幫助人們了解意見領袖的聲量是如何累積，以及他們是如何讓其關注的議題被更多人看見，最後，我們會做使用者研究來評估本論文的成效。

關鍵字

視覺化 ； 社群媒體 ； 使用者 ； 行為 ； 探索 ； 系統

並列摘要

Social media is one of the main channels for obtaining information in modern society, and it is also a place where people can express their opinions. However, people's views are often influenced by opinion leaders on social media. The influence of opinion leaders is usually accumulated by publishing articles and comments, but it may also be carried by specific groups. Therefore, this research proposes a visual analysis system, which expresses the behavior similarity between users as a sorted distance matrix, and then visualizes them based on an ordered heatmap. We also use bipartite relations between users and articles to draw it into visualization with an adjacency matrix style. By combining the above two visualizations, we can observe users' activities and compare the differences between different types of users. The system can help people understand how to accumulate the influence of the opinion leader and how they make their concerns more visible.

並列關鍵字

Visual ； Analytic ； System ； Social media ； User ； Behavior

參考文獻

[1] J. Heer, & D. Boyd. Vizster: visualizing online social networks, IEEE Symposium on Information Visualization (INFOVIS 2005), 2005.

[2] Nan Cao, Yu-Ru Lin, Fan Du, & Dashun Wang. Episogram: Visual Summarization of Egocentric Social Interactions. IEEE Computer Graphics and Applications 36.5: 72-81, 2016.

[3] Mengdie Hu, Krist Wongsuphasawat, & John Stasko. Visualizing Social Media Content with SentenTree. IEEE Transactions on Visualization and Computer Graphics, 23.1, 2017.

[4] P. Xu, Y. Wu, E. Wei, T.-Q. Peng, S. Liu, J. J. H. Zhu, & H. Qu. Visual analysis of topic competition on social media. IEEE Transactions on Visualization and Computer Graphics, 19.12:2012–2021, 2013.

[5] F. Viégas, M. Wattenberg, & J. Hebert. Google+Ripples: a native visualization of information flow. WWW '13: Proceedings of the 22nd international conference on World Wide Web, pages 1389-1398, 2013.