簡易檢索 / 詳目顯示

回結果列表
研究生： 蔡佳芸
JIA-YUN TSAI
論文名稱： 透明度至關重要：深入探討於教育的可解釋機器學習 — 有風險學習學生的辨識應用
Transparency Matters: A Dive into Explainable Machine Learning for Education with a Focus on Identifying At-Risk Students
指導教授： 呂志豪
Shih-Hao Lu
口試委員: 黃政嘉
Jheng-Jia Huang
黃振皓
Chen-Hao Huang
學位類別： 碩士
Master
系所名稱： 管理學院 - 企業管理系
Department of Business Administration
論文出版年： 2023
畢業學年度： 112
語文別： 英文
論文頁數： 38
中文關鍵詞： 教育數據探勘 、學習風險預測 、機器學習 、可解釋機器學習
外文關鍵詞： Educational Data Mining, At-risk student Prediction, Machine Learning, Explainable Machine Learning
相關次數： 點閱：71  下載：1 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 近年來因為疫情的影響，使得數位化教育能在小學與國中實行。以前許多文獻研究的對象都為大學學生的學習情境，所以拓展小學與國中數位化的相關研究至關重要。數位化教育能帶來許多正向的改變，以提升教學機構的效率與學生學習的效率。
    本研究聚焦小學線上與線下整合資料，除了用機器學習模型精準的預測有風險學習者的學習結果，亦更近一步的運用可解釋機器學習模型探討學習行為與學習結果之間的關係併且用全域與局部的角度來解釋模型。
    這項研究指出歷史學習成績在預測未來表現上扮演重要角色，但同時也顯示成績可透過個人努力而改善。因此，我們強調線上學習行為相關的變數具有相當的重要性，這有助於辨識與「需注意」和「表現不及格」的學生高度相關的行為模式，進而採取預防性措施。
    這項研究在準確預測「需注意」和「表現不及格」的學生方面做出了顯著的貢獻。更為重要的是，它深入探討影響這些預測的因素，使我們能夠揭示最具影響力的變數。識別這些關鍵因素對於推薦有針對性的干預措施以提升學術表現至關重要。在探索未知領域的過程中，這項研究為塑造個性化教育未來的預測模型帶來了新的光明。

    In the ever-evolving landscape of education, the intersection of technology and Educational Data Mining has given rise to transformative methodologies. Most studies focus on the accuracy of predicting at-risk students, but few focus on model transparency. This study delves into Educational Data Mining, where machine learning and Explainable Machine Learning (XML) converge to identify at-risk students.
    The methodology focuses on integrating online and offline data in primary education to actively predict the learner's concerns and underperformance and improve the learner's performance. In addition, XML provides insights into the “black box” of complex algorithms, empowering users to comprehend and validate model outputs. By explaining how the model predicts results from a global and local perspective, we can understand how the model predicts results.
    This study found the significant role of historical academic performance in predicting future outcomes while highlighting the potential for improvement through other efforts. Consequently, we underscore the importance of variables associated with online learning behaviors. These factors can aid in identifying behavior patterns strongly correlated with students requiring attention or exhibiting poor performance, facilitating the implementation of intermediation.
    This study makes significant contributions by accurately predicting concerns and underperformance of students. More crucially, it delves into the factors influencing these predictions, allowing us to unveil the most impactful feature. Identifying these key factors is pivotal for recommending targeted interventions to enhance academic performance. In navigating unexplored realms, this research sheds light on the predictive models that will shape the future of personalized education.

    TABLE OF CONTENTS 摘要 I ABSTRACT II 誌謝 III TABLE OF CONTENTS IV LIST OF TABLES VI LIST OF FIGURES VI CHAPTER 1 INTRODUCTION 1 1.1 Research Background 1 1.2 Research Motivation 4 1.3 Research Objectives 5 1.4 The Importance of Research 5 CHAPTER 2 LITERATURE REVIEW 7 2.1 Educational Data Mining In Academic Progress 7 2.2 Multiple Data Sources 8 2.3 ML Models for Student Performance Prediction 9 2.4 An Intervention Measurement 11 CHAPTER 3 RESEARCH METHODOLOGY 13 3.1 Dataset Description 13 3.2 Data Preprocessing And Feature Engineering 15 3.3 Interpretable ML Model 17 3.4 Research Design 19 CHAPTER 4 EXPERIMENTAL RESULTS 21 4.1 Statistically Feature Selection 21 4.2 ML Model Performance 22 4.3 The Insights From the XML Method 23 4.3.1 Interpret The Feature Importance By SHAP Value 24 4.3.2 A Deep Dive into XML Using Beeswarm Plots 25 4.3.3 Simulation Model Results through Partial Dependence Plots 28 4.3.4 A Visual Breakdown of LIME Insights 31 CHAPTER 5 CONCLUSIONS AND FUTURE RESEARCH 33 5.1 Conclusions and Contribution 33 5.2 The Limitation and Future Research 35 REFERENCES 36

    Adnan, M., Habib, A., Ashraf, J., Mussadiq, S., Raza, A. A., Abid, M., Bashir, M., & Khan, S. U. (2021). Predicting at-Risk Students at Different Percentages of Course Length for Early Intervention Using Machine Learning Models. IEEE Access, 9, 7519–7539. https://doi.org/10.1109/ACCESS.2021.3049446
    Albreiki, B., Habuza, T., & Zaki, N. (2022). Framework for automatically suggesting remedial actions to help students at risk based on explainable ML and rule-based models. International Journal of Educational Technology in Higher Education, 19(1). https://doi.org/10.1186/s41239-022-00354-6
    Ameloot, E., Rotsaert, T., & Schellens, T. (2021). The supporting role of learning analytics for a blended learning environment: Exploring students’ perceptions and the impact on relatedness. Journal of Computer Assisted Learning. https://doi.org/10.1111/jcal.12593
    Azcona, D., Hsiao, I-Han., & Smeaton, A. F. (2019). Detecting students-at-risk in computer programming classes with learning analytics from students’ digital footprints. User Modeling and User-Adapted Interaction, 29(4), 759–788. https://doi.org/10.1007/s11257-019-09234-7
    Baek, C., & Doleck, T. (2021). Educational Data Mining versus Learning Analytics: A Review of Publications From 2015 to 2019. Interactive Learning Environments, 31(6), 3828–3850. https://doi.org/10.1080/10494820.2021.1943689
    Chen, H.-L. S., & Yu, P. (2016). Closing achievement gaps and beyond: teachers’ reactions to the remedial education policy in Taiwan. Asia Pacific Education Review, 17(4), 609–624. https://doi.org/10.1007/s12564-016-9440-y
    Geraci, L., Kurpad, N., Tirso, R., Gray, K. N., & Wang, Y. (2022). Metacognitive errors in the classroom: The role of variability of past performance on exam prediction accuracy. Metacognition and Learning. https://doi.org/10.1007/s11409-022-09326-7
    Guerrero-Roldán, A.-E., Rodríguez-González, M. E., Bañeres, D., Elasri-Ejjaberi, A., & Cortadas, P. (2021). Experiences in the use of an adaptive intelligent system to enhance online learners’ performance: a case study in Economics and Business courses. International Journal of Educational Technology in Higher Education, 18(1). https://doi.org/10.1186/s41239-021-00271-0
    Guo, Y. M., Klein, B. D., & Ro, Y. K. (2019). On the effects of student interest, self-efficacy, and perceptions of the instructor on flow, satisfaction, and learning outcomes. Studies in Higher Education, 45(7), 1413–1430. https://doi.org/10.1080/03075079.2019.1593348
    Hongsuchon, T., Emary, I. M. M. E., Hariguna, T., & Qhal, E. M. A. (2022). Assessing the Impact of Online-Learning Effectiveness and Benefits in Knowledge Management, the Antecedent of Online-Learning Strategies and Motivations: An Empirical Study. Sustainability, 14(5), 2570. https://doi.org/10.3390/su14052570
    Jayawardena, P. R., van Kraayenoord, C. E., & Carroll, A. (2020). Factors that influence senior secondary school students’ science learning. International Journal of Educational Research, 100, 101523. https://doi.org/10.1016/j.ijer.2019.101523
    Kim, W.-H., & Kim, J.-H. (2020). Individualized AI Tutor Based on Developmental Learning Networks. IEEE Access, 8, 27927–27937. https://doi.org/10.1109/ACCESS.2020.2972167
    Kurilovas, E. (2018). Advanced machine learning approaches to personalise learning: learning analytics and decision making. Behaviour & Information Technology, 38(4), 410–421. https://doi.org/10.1080/0144929x.2018.1539517
    Luan, H., & Tsai, C.-C. (2021). A Review of Using Machine Learning Approaches for Precision Education. Educational Technology & Society, 24(1), 1176–3647.
    Ninaus, M., Greipl, S., Kiili, K., Lindstedt, A., Huber, S., Klein, E., Karnath, H.-O., & Moeller, K. (2019). Increased emotional engagement in game-based learning – A machine learning approach on facial emotion detection data. Computers & Education, 142, 103641. https://doi.org/10.1016/j.compedu.2019.103641
    Pek, R. Z., Özyer, S. T., Elhage, T., ÖZYER, T., & Alhajj, R. (2023). The Role of Machine Learning in Identifying Students At-Risk and Minimizing Failure. IEEE Access, 11, 1224–1243. https://doi.org/10.1109/ACCESS.2022.3232984
    Sghir, N., Adadi, A., & Lahmer, M. (2022). Recent advances in Predictive Learning Analytics: A decade systematic review (2012–2022). Education and Information Technologies. https://doi.org/10.1007/s10639-022-11536-0
    Singh, H., Kaur, B., Sharma, A., & Singh, R. (2023). Framework for suggesting corrective actions to help students intended at risk of low performance based on experimental study of college students using explainable machine learning model. Education and Information Technologies. https://doi.org/10.1007/s10639-023-12072-1
    Tsai, S.-C., Chen, C.-H., Shiao, Y.-T., Ciou, J.-S., & Wu, T.-N. (2020). Precision education with statistical learning and deep learning: a case study in Taiwan. International Journal of Educational Technology in Higher Education, 17(1). https://doi.org/10.1186/s41239-020-00186-2

    QR CODE