資料視覺化分析對於資料處理、探索資料結構、辨識趨勢及叢聚、發現局部模式（pattern or subgroup）、評估模型分析輸出（output）與呈現分析結果都相當有幫助，對於探索性數據分析（exploratory data analysis）、網絡分析（network analysis）及機器學習等大數據分析更是不可或缺。然而資料科學成功的基礎為特徵工程（Feature Engineering），它是資料科學流程中最耗費時間的步驟，包含準備階段、生成階段、轉換階段、建模階段及操作階段等流程。

本課程主要目標為使用SAS及R軟體與實際調查資料，來展現資料視覺化分析能展現資料中的哪些資訊。透過實際操作來增加具體設計、分析及詮釋統計圖像的經驗，來有效率地了解資料視覺化分析與機器學習。

Data visual analysis is of great help in data processing, exploring data structure, identifying trends and aggregation, discovering local patterns or subgroups, evaluating model analysis output (output) and presenting analysis results, and is also indispensable for large data analysis such as exploratory data analysis, network analysis and machine learning. However, the foundation of successful data science is Feature Engineering, which is the most time-consuming step in the data science process, including preparation stages, generation stages, conversion stages, modeling stages, and operation stages.

The main purpose of this course is to use SAS and R software and actual survey data to show what information can be displayed in the data visual analysis. Through actual operation, add experience in design, analysis and simplify statistical images to effectively understand data visual analysis and machine learning.

1. Exploring SAS® Viya®: Visual Analytics, Statistics, and Investigations
2. 沈葆聖(2002)，SAS統計軟體與資料分析(滄海書局)

1. Exploring SAS® Viya®: Visual Analytics, Statistics, and Investigations
2. Shen Baosheng (2002), SAS Statistical Software and Data Analysis (Yanghai Books Bureau)