對於線上教學平台學習行為的分析逐漸變成重要的環節，除了能讓授課教師從學習分析的結果來觀察學生的學習進度與情況，透過學習分析及時發現進度落後的學生且給予協助；透過學習分析套件得到的結論，促使授課教師調整課程難度與進度，在 OpenEdu 平台有兩種分析系統一種是 Insights，另一種是 Dashboard 這兩種系統，並且兩者都有共通點，其共通點為系統內提供的分析都為固定的對於想要知道更詳細內容或是客製化的內容是有難度的。

為了解決以上問題，本論文提出了學習行為分析套件 (PyOpenEdu) 的框架，並且框架是藉由物件導向設計而設計出來，依據功能架構建立基本的物件、設計每個物件之功能與設計樣式設計框架讓框架更加完整，透過發展出一個 PyOpenEdu 框架的學習分析流程使得使用者降低了學習分析所需時間與程式碼的複雜度，同時也讓學習分析變得更容易使用。

Analysis of learning behavior on online teaching platforms has gradually become important. In addition to allowing teachers to observe the learning progress and conditions of students from the results of learning analysis, they can assist students who are behind the learning progress. The conclusions obtained through the learning analysis package prompt the instructor to adjust the difficulty and progress of the course There are two analysis systems on the OpenEdu platform, one is Insights and the other is Dashboard, and both have common points. The common points are provided by the system. The analysis is fixed and it is difficult to know more detailed content or customized content.

In order to solve the above problems, this paper proposes a framework of the learning behavior analysis suite (PyOpenEdu) and the framework is designed by object-oriented design, based on the functional framework to establish basic objects, design each object's function and design style design framework. In order to make the framework more complete, by developing the learning analysis process of the PyOpenEdu framework, users can reduce the time required for learning analysis and the complexity of the code, while also making learning analysis easier to use.

目錄
摘要 1
Abstract 2
目錄 3
圖目錄 5
表目錄 6
第一章、緒論 7
1.1 研究背景 7
1.2 研究動機 7
1.3 研究目的 8
1.4 章節介紹 8
第二章、相關技術與研究 9
2.1 相關技術 9
2.1.1 軟體框架 9
2.1.2 學習分析技術 10
2.2 相關研究 11
2.2.1 學習科技相關的研究 11
2.2.2 軟體框架相關研究 13
第三章、研究方法 15
3.1 PyOpenEdu 15
3.2 學習分析 16
3.2.1 描述性分析 16
3.2.2 統計性分析 17
3.2.3 預測性分析 18
3.3 框架的擴充性 19
3.4 PyOpenEdu 框架的學習分析流程 20
第四章、實驗 22
4.1 Python 與資料科學入門 22
4.2 定義問題目標與資料準備 22
4.3 資料前處理與匯總資料 23
4.3.1 學生歷程資料前處理與匯總 24
4.3.2 OJ 測驗題資料擴充資料前處理與匯總 26
4.4 學習分析與結果 27
4.4.1 分析問題一 28
4.4.2 分析問題二 29
第五章、結論 32
參考文獻 33

[1] Fayad, Mohamed, and Douglas C. Schmidt. "Object-oriented application frameworks."
Communications of the ACM 40.10 (1997): 32-38.
[2] Márquez-Vera, Carlos, Cristóbal Romero Morales, and Sebastián Ventura Soto.
"Predicting school failure and dropout by using data mining techniques." IEEE Revista
Iberoamericana de Tecnologias del Aprendizaje 8.1 (2013): 7-14.
[3] Yang, Stephen JH, et al. "Predicting students' academic performance using multiple
linear regression and principal component analysis." Journal of Information Processing
26 (2018): 170-176.
[4] Rajni, Jindal, and Dutta Borah Malaya. "Predictive analytics in a higher education
context." IT Professional 17.4 (2015): 24-33.
[5] Zorrilla, Marta Elena, and Elena Álvarez. "MATEP: Monitoring and analysis tool for
e-learning platforms." 2008 Eighth IEEE International Conference on Advanced
Learning Technologies. IEEE, 2008.
[6] Pardos, Zachary A., Anthony Whyte, and Kevin Kao. "moocRP: Enabling open
learning analytics with an open source platform for data distribution, analysis, and
visualization." Technology, Knowledge and Learning 21.1 (2016): 75-98.
[7] Solar:
https://www.solaresearch.org/about/what-is-learning-analytics/