資料分析於輔助線上學習的即時線上考試系統

Real-time Online Assessment with Data Analyze System to Support E-Learning

蘇冠中(Su Guan-Chung)

指導教授 ： 賀耀華

國立臺灣師範大學/理學院/資訊工程學系/碩士(2022年)

https://doi.org/10.6345/NTNU202200562

摘要

由於 2019 年的 COVID-19 傳染病大流行的緣故，高等教育在一夜之間面臨 需要採用線上課程的方式來進行授課。在線上課程中，教學方式、考試評量與學 習評估等都與實體課程有非常大的差異，由於線上考試必須得使用電腦作答，但 作答電腦本身具有上網的功能，學生在考試期間隨時都可以透過網路去搜尋答案， 因此較難對學生的學習成效進行評估，教師也難以確認課程的教學成效。
在本篇研究中提出資料分析於輔助線上學習的即時線上考試系統(Real-time Online Assessment with Data Analyze System to Support E-Learning, ROAD)。首先 ROAD 可將一般題目進行隨機化讓每位考生的題目順序不同，並設置作答完無法 返回與較短的時間限制，轉變為線上考試的設置，以降低學生互相交流答案的可 能性;接著在線上考試中，ROAD 能夠紀錄學生在考試期間電腦上的行為模式， 並辨識可疑的作弊行為;最後通過決策樹演算法分析學生的線上行為模式資料與 考試的分數結果，我們能給予教師及學生回饋，讓師生瞭解該堂課的學習成效， 並幫助教師針對課程教學進行改善。

關鍵字

線上考試 ； 數位學習 ； 教學評估 ； 行為模式 ； 決策樹

並列摘要

Due to the sudden outbreak of the COVID-19 in 2019, higher education institutions are facing the need to adopt online teaching. Online courses are very different from physical courses with teaching methods, test evaluation, and learning assessments. Considering that online assessment is usually done on a computer, students can search the answers from the internet during the assessment. Therefore, it is difficult to evaluate the effectiveness of both students and teachers for the online courses.
In this research, we proposed a Real-time Online Assessment with Data Analyze System to Support E-Learning (ROAD) to assist online learning assessment. First, ROAD can randomize general questions so that each student has a different sequential order of viewing questions, and set that cannot be returned after answering with shorter time limit, by these three part can reduce the possibility of students exchanging answers with each other and turn the questions into online assessments. Second, ROAD can record students on computer’s behavior patterns during the assessment and identify suspicious cheating behaviors. Finally, by analyzing student’s behavior patterns with online assessment score results through decision tree algorithm, we can give teachers and students feedback to improve the effectiveness of learning and teaching for the online courses.

並列關鍵字

Online Assessment ； E-learning ； Instructional Evaluation ； behavior patterns ； Decision tree

參考文獻

[1] Garg, K., Verma, K., Patidar, K., & Tejra, N. (2020, May). Convolutional neural network based virtual exam controller. In 2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS) (pp. 895-899). IEEE.

[2] Clark, T. M., Callam, C. S., Paul, N. M., Stoltzfus, M. W., & Turner, D. (2020). Testing in the time of COVID-19: A sudden transition to unproctored online exams. Journal of chemical education, 97(9), 3413-3417.

[3] Chua, S. S., Bondad, J. B., Lumapas, Z. R., & Garcia, J. D. L. (2019, October). Online examination system with cheating prevention using question bank randomization and tab locking. In2019 4th International Conference on Information Technology (InCIT) (pp. 126-131). IEEE.

[4] Jalali, K., & Noorbehbahani, F. (2017). An automatic method for cheating detection in online exams by processing the student’s webcam images. In Proc. 3rd Conf. Elect. Comput. Eng. Technol.(E-Tech) (pp. 1-6).

[5] Shruti, M. I., Pratiksha, M. K., Jyoti, M. M., Snehal, M. N., & Vaibhav, D. Designing Security Framework for Secure Exam System based on QR Code.