隨著資訊技術、社交媒體和感測器等科技之爆炸性發展，資料儲存量的積累正以隨機或有組織之方式，呈指數曲線之速度快速增長。「大數據」成為近年新穎的話題之一，吸引大量領域使用人工智慧、深度學習等創新技術來探索隱含於巨量資料中的新知識或新資訊。視覺化為大數據分析步驟中的一環，基於大量資訊具有時間及空間之特性，如何強化地理資訊設計及處理之策略及地理視覺化機制之發展，以有效突顯與強化其時空之面向，是擴充地理資訊智慧應用的關鍵挑戰。本研究提出基於Cube之地理資料組織架構，將地理資料之時間、空間和變量等三個主要面向分別建構為Cube之三個軸向，三軸依據屬性多寡或是資料記錄之複雜程度具有不同的細節，任何地理資料之特性均可以Cube之單一元素代表，並可基於三軸之操作特性而組合或拆解單一元素中之資料，進而形成另一個特性組合之資料。意味在Cube之架構中，原始資料將可基於發展之知識而產生更多元的資料內容，進而活化Cube在決策與分析的可能性。Cube中之單一元素具有其各自的假設與要求，因此適用於特定之視覺化方法。本研究歸納與整理各類視覺化技術與先決條件後，透過雙向分析，建議各資料組合適用之視覺化技術。為保證地理資料特性、資料組合與製圖技術之正確定義及分類，本研究也針對可成功達成地理視覺化之品質條件提出品質檢核項目之建議，以確保視覺化成果之正確性基於提出之架構可能產生大量數目的視覺化成果，本研究進一步納入基於領域或統計知識之門檻機制，自動化篩選與過濾具有顯著空間分布、時序演進或統計樣式之成果，最後由發展之視覺化機制中選擇最適當之方式提示。

以近年之發展趨勢而言，領域資料數量無疑將持續成長，其中也無庸置疑隱藏著高度的應用價值，但缺乏有效之管理、分析與視覺化對策將造成大量、重負起甚至沒有必要的處理工作，嚴重阻礙大數據的應用。本研究著眼於有效歸納地理資料之特性，並藉由管理及知識機制之納入而挖掘出其中值得決策者關注之資訊，最後再以合適之視覺化技術展現。智慧化機制之優勢在於減低決策者在資料操作與展示上的技術門檻，並可達到活用跨域或時序性資料，進而彈性提供多元產品之目標。有鑑於目前之視覺化軟體工具多聚焦在簡化功能之操作，但鮮少著墨於結合地理資料之知識及自動化分析判斷之需求，本研究之成果可提供視覺展示軟體與大數據資料庫有效結合之發展基礎，並具有進一步帶動地理資訊軟體進化的潛力。

With the breakthrough of information technology, social media and sensing technology, a large amount of data has accumulated rapidly in a random or organized way. "Big Data" has received tremendous attention in recent years. People use many innovative technologies, such as artificial intelligence and deep learning, to explore new knowledge or new information hidden in the increasingly huge volume of data. Visualization is one of the steps of big data analysis. The data in real world has two major consideration, "space" and "time". How to highlight these two aspects in the development of geographic information design and processing, so as to explore the smart applications based geographic information will be an important challenge to the future research. The research purposes a cube-based concept of geographic data organization. Three major characteristics of geographic data, namely temporal, spatial and variables, are selected as the three major modelling perspective of the cube. Each axis has different details based on the numbers of attributes or the complexity of their respective data recording methods. Each geographic data can be classified and modelled by a single cube element based on its distinguished characteristics. According to the hierarchical design, cube elements can be consolidated or disassemble to generate new information following specified rules of data quality. As every cube element is defined with a unique set of distinguished geographic characteristic, it has specific assumption, requirement and corresponding visualization methods. After summarizing various prerequisites of visualization methods, this research suggests the appropriate visualization methods for each cube element through two-way analysis. To ensure the correct definition and classification of geographic data characteristics, data combination and visualization methods, this research also proposes quality constraints for the processing of geographic data. This can smartly select the appropriate visualization strategies and ensure the correctness of visualized results.

The recent trend undoubtedly indicates there is a high application value hidden in the big data. However, the lack of effective management, analysis, and visualization strategies will cause heavy and unnecessary processing work. It seriously hinders the application of big data. This research focuses on the effective induction of the characteristics of geographic data, using management and knowledge mechanisms to explore hidden information, and finally displaying the information with appropriate visualization techniques to decision makers. The advantage of this intelligent mechanism is to bridge the gap between decision-makers and the huge volume of unfamiliar domain data, so as to achieve the goal of cross-domain and time series data interoperability for enabling flexible and multi-purpose applications. While the current visualization software mainly focuses on simplifying the operation of functions and seldomly focuses on the combination of geographic data knowledge, automatic analysis and judgment, this research provides a solid foundation for the effective combination of visualization software and big data database, and have the potential to drive the evolution of geographic information software.

公路路線系統分類基準(2014). 交通部。
邱皓政，溫福星.(2007). 脈絡效果的階層線性模型分析：以學校組織創新氣氛與教師創意表現為例. 國立政治大學「教育與心理研究」學刊，30(1)，1-35。
國土資訊系統跨域資料品質共同規範(2019). 內政部資訊中心。
Abela, A. (2009). Chart Suggestions—A Thought-Starter.
Ackoff, R. L. (1989). From Data to Wisdom, Journal of Applies Systems Analysis, 16, 3-9.
Borner, K. (2015). Atlas of Knowledge: Anyone can map.
Card, S. K., Mackinlay, J. D. et al., (1999). Readings in Information Visualization; Using Vision to think. Academic Press.
Dykes, J. et al. (2005). Exploring Geovisualization, United Kingdom: Elsevier Ltd..
Friendly (2006). A Brief History of Data Visualization.
Gandomi, A. & Haider, M. (2015). Beyond the Type: Big data concepts, methods, and analytics. International Journal of Information Management. 35(2), 137–144.
Hajirahimova1&Ismayilova (2018). BIG DATA VISUALIZATION: EXISTING APPROACHES AND PROBLEMS，Problems of information technology, 1, 65-74.
Keim, D.A. and Kriegel, H. P. (1996). Visualization techniques for mining large databases: a comparison, IEEE Transactions on Knowledge and Data Engineering, 8(6), 923-936.
Keim, D.A. (2002). Information visualization and visual data mining, IEEE Transactions on Visualization and Computer Graphics, 8(1),1-8.
Khan, M., and Khan, S. S. (2011). Data and information visualization methods, and interactive mechanisms: A survey. International Journal of Computer Applications, 34(1), 1-14.
Laney, D., 2001. 3D Data Management: Controlling Data Volume, Velocity and Variety.
Luo, Y., Qin, X., Tang, N. and Li, G. (2018). DeepEye: Towards Automatic Data Visualization, IEEE 34th International Conference on Data Engineering (ICDE), 101-112.
MacEachren, A. M., & Kraak, M. J. (1997). Exploratory cartographic visualization advancing the agenda. Computers & geosciences, 23(4), 335-343.
Mackinlay, J., Hanrahan, P., Stolte, C. (2007). Show Me: Automatic Presentation for Visual Analysis, IEEE Transactions on Visualization and Computer Graphics, 13(6), 1137 – 1144.
Olshannikova, E., A. Ometov, Koucheryavy, Y. and Olsson, T. (2015). Visualizing Big Data with augmented and virtual reality: challenges and research agenda, Journal of Big Data, 2(22).
Rahm, E. & Do, H.H. (2000). Data Cleaning: Problems and Current Approaches. Bulletin of the Technical Committee on Data Engineering, 23(4), 3-13.
Raper, J. et al. (2000). Multidimensional Geographic Information Science. London, England: Taylor & Francis Group.
Shneiderman, B. (1996). The eyes have it: a task by data type taxonomy for information visualizations, Proceedings 1996 IEEE Symposium on Visual Languages, Boulder, CO, USA, 336-343.
Sun Y., Leigh J., Johnson A. and Lee S. (2010). Articulate: A Semi-automated Model for Translating Natural Language Queries into Meaningful Visualizations. Taylor R., Boulanger P., Krüger A., Olivier P. (eds) Smart Graphics. SG 2010. Lecture Notes in Computer Science, vol 6133. Springer, Berlin, Heidelberg
Uthayasankar Sivarajah, Muhammad Mustafa Kamal, Zahir Irani, Vishanth Weerakkody (2017). Critical analysis of Big Data challenges and analytical methods, Journal of Business Research, 70, 263-286,
Wills, G. and Wilkinson, L. (2010). AutoVis: Automatic Visualization. Information Visualization, 9(1), 47-69.