促進學習體驗的研究在近期有增加的趨勢，其中基於學習風格的個體差異也被廣泛討論，希望透過學習風格的辨識增進學習效率。於此研究中我們主要探討VARK學習風格，相較於原始的四個類別(視覺型、聽覺型、讀寫型、動覺型)，我們僅採用線上課程中常見的三種學習內容呈現方式，視覺、聽覺以及文字(讀寫)，以此貼近線上的環境。相較於傳統的學習風格辨識方法(問卷)，我們旨在利用腦電波提出一個較為客觀的方法進行學習風格辨識。大腦與學習任務之間有著高度的相關性，在科技的進步下，腦部活動可以藉由非侵入式的方式測量，腦電波為其中之一，基於此我們使用腦電波作為分析學習風格差異之依據，我們記錄受試者在觀看不同種類的教學素材時的腦電波資料，並使用孿生網路架構的分類器緩解資料不平衡所造成的實驗結果偏差。在留一受試者的交叉驗證方法下達到宏觀(macro)平均f1分數0.41的VAR(視覺型、聽覺型、讀寫型)學習風格分類結果，並在聽覺型、視覺型、讀寫型學習風格的二分類任務中分別最高獲得f1分數0.70、0.54以及0.41的分類結果。在此研究中，我們並不僅提出一個較為客觀的學習風格辨識方法可以做為日後學習風格辨識的輔助工具，並展示了腦電波成為學習風格分類依據的潛力，作為日後研究的參考。

Individual difference in learning was widely studied in education, and learning style is one of the concepts that can affect learning efficiency. In this study, we used the VARK learning style model including the Visual category, the Aural category, the Reading/writing category, and the Kinesthetic category. Instead of the four categories proposed in the original concept, we only focused on VAR (Visual, Aural, Reading/Writing) three categories which are the common representation of learning materials in the online environment. We aimed to provide a more objective method for learning style recognition to support the conventional questionnaires. Considering the correlation between the brain and learning, we explored the relationship between the brain activity of learners and their learning styles. Benefiting from the technique improvement, brain activity can be measured by non-invasive methods. Electroencephalography (EEG) was adopted in our experiment and we recorded the EEG signals of subjects during the different learning-related materials presenting. To tackle the data imbalance problem, the siamese neural network was employed to recognize learning styles. We achieved a macro averaged f1-score of 0.41 based on leave-one-subject-out testing by adopting the siamese neural network. In addition to the multilabel learning style recognition, we conducted three binary discrimination tasks for all three learning style recognition, and f1-scores of 0.70, 0.54, and 0.41 were achieved for the aural learning style, visual learning style, and reading/writing learning style discrimination tasks, respectively.
In this study, we showed that EEG could potentially be a more objective tool for learning style recognition which can support the conventional questionnaire. We also explored the related EEG features which were contributed to learning style recognition, and the results can be referred to for the follow-up research.

Abstract (Chinese) ---------- I
Acknowledgements (Chinese) -- II
Abstract -------------------- III
Acknowledgements ------------ IV
Contents -------------------- V
List of Figures ------------- VIII
List of Tables -------------- X
1 Introduction -------------- 1
2 Methodology --------------- 15
3 Result -------------------- 36
4 Discussion ---------------- 58
5 Conclusion ---------------- 69
6 Supplementary ------------- 79

[1] Hui Luan and Chin-Chung Tsai. A review of using machine learning approaches for precision education. Educational Technology & Society, 24(1):250–266, 2021.
[2] Michal Kr ́ol and Magdalena Kr ́ol. Learning from peers’ eye movements in the absence of expert guidance: A proof of concept using laboratory stock trading, eye tracking, and machine learning. Cognitive science, 43(2):e12716, 2019.
[3] Wanli Xing and Dongping Du. Dropout prediction in moocs: Using deep learning for personalized intervention. Journal of Educational Computing Research, 57(3):547–570, 2019.
[4] Frank Romanelli, Eleanora Bird, and Melody Ryan. Learning styles: a review of theory, application, and best practices. American journal of pharmaceutical education, 73(1), 2009.
[5] Neil D Fleming and Colleen Mills. Not another inventory, rather a catalyst for reflection. To improve the academy, 11(1):137–155, 1992.
[6] Simon Cassidy*. Learning styles: An overview of theories, models, and measures. Educational psychology, 24(4):419–444, 2004.
[7] Milan Klement. How do my students study? an analysis of students’ of educational disciplines favorite learning styles according to vark classification. Procedia-Social and Behavioral Sciences, 132:384–390, 2014.
[8] Haohan Wang, Yiwei Li, Xiaobo Hu, Yucong Yang, Zhu Meng, and Kai-min Chang. Using eeg to improve massive open online courses feedback interaction. In AIED Workshops.
[9] Fu-Ren Lin and Chien-Min Kao. Mental effort detection using eeg data in e-learning contexts. Computers & Education, 122:63–79, 2018.
[10] Lukas AW Gemein, Robin T Schirrmeister, Patryk Chrabaszcz, Daniel Wilson, Joschka Boedecker, Andreas Schulze-Bonhage, Frank Hutter, and Tonio Ball. Machine-learning-based diagnostics of eeg pathology. NeuroImage, 220:117021, 2020.
[11] Marzieh Savadkoohi, Timothy Oladunni, and Lara Thompson. A machine learning approach to epileptic seizure prediction using electroencephalogram (eeg) signal. Biocybernetics and Biomedical Engineering, 40(3):1328–1341, 2020.
[12] Yueer Wei, Qingxia Yang, Jiangping Chen, and Jie Hu. The exploration of a machine learning approach for the assessment of learning styles changes. Mech. Syst. Control Formerly Control Intell. Syst., 46(3):21, 2018.
[13] Bryan Kolb. Brain plasticity and behavior. Psychology Press, 2013.
[14] Usha Goswami. Neuroscience and education. British journal of Educational psychology, 74(1):1–14, 2004.
[15] Bruce D McCandliss, Michael I Posner, and Thomas Givon. Brain plasticity in learning visual words. Cognitive Psychology, 33(1):88–110, 1997.
[16] Joenna Driemeyer, Janina Boyke, Christian Gaser, Christian B ̈uchel, and Arne May. Changes in gray matter induced by learning—revisited. PloS one, 3(7):e2669, 2008.
[17] C`eline Chayer and Morris Freedman. Frontal lobe functions. Current neurology and neuroscience reports, 1(6):547–552, 2001.
[18] James W Bisley and Michael E Goldberg. Attention, intention, and priority in the parietal lobe. Annual review of neuroscience, 33:1, 2010.
[19] Ingrid R Olson and Marian Berryhill. Some surprising findings on the involvement of the parietal lobe in human memory. Neurobiology of learning and memory, 91(2):155–165, 2009.
[20] Khald Ali I Aboalayon, Miad Faezipour, Wafaa S Almuhammadi, and Saeid Moslehpour. Sleep stage classification using eeg signal analysis: a comprehensive survey and new investigation. Entropy, 18(9):272, 2016.
[21] Alexander A Borb ́ely, Fritz Baumann, Daniel Brandeis, Inge Strauch, and Dietrich Lehmann. Sleep deprivation: effect on sleep stages and eeg power density in man. Electroencephalography and clinical neurophysiology, 51(5):483– 493, 1981.
[22] Xiao-Wei Wang, Dan Nie, and Bao-Liang Lu. Emotional state classification from eeg data using machine learning approach. Neurocomputing, 129:94–106, 2014.
[23] Y. Zhou, T. Xu, S. Li, and S. Li. Confusion state induction and eeg-based detection in learning. In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pages 3290–3293.
[24] Paula D Ladd and Ralph Ruby Jr. Learning style and adjustment issues of international students. Journal of education for business, 74(6):363–367, 1999.
[25] Rita Dunn. Learning style: State of the science. Theory into practice, 23(1):10–19, 1984.
[26] Vanessa Marcy. Adult learning styles: How the vark learning style inventory can be used to improve student learning. Perspectives on Physician Assistant Education, 12(2):117–120, 2001.
[27] Saul McLeod. Kolb’s learning styles and experiential learning cycle. Simply psychology, 2017.
[28] Bernice McCarthy. Using the 4mat system to bring learning styles to schools. Educational leadership, 48(2):31–37, 1990.
[29] Richard M Felder. Matters of style. ASEE prism, 6(4):18–23, 1996.
[30] Richard M Felder, Barbara A Soloman, et al. Learning styles and strategies, 2000.
[31] Jane Bromley, Isabelle Guyon, Yann LeCun, Eduard S ̈ackinger, and Roopak Shah. Signature verification using a” siamese” time delay neural network. Advances in neural information processing systems, 6, 1993.
[32] Mohsen Heidari and Kazim Fouladi-Ghaleh. Using siamese networks with transfer learning for face recognition on small-samples datasets. In 2020 International Conference on Machine Vision and Image Processing (MVIP), pages 1–4, 2020.
[33] N. S. Suhaimi, J. Mountstephens, and J. Teo. Eeg-based emotion recognition: A state-of-the-art review of current trends and opportunities. Comput Intell Neurosci, 2020:8875426, 2020.
[34] Obada Al Zoubi, Chung Ki Wong, Rayus T Kuplicki, Hung-wen Yeh, Ahmad Mayeli, Hazem Refai, Martin Paulus, and Jerzy Bodurka. Predicting age from brain eeg signals—a machine learning approach. Frontiers in aging neuroscience, 10:184, 2018.
[35] Soyiba Jawed, Hafeez Ullah Amin, Aamir Saeed Malik, and Ibrahima Faye. Classification of visual and non-visual learners using electroencephalographic alpha and gamma activities. Frontiers in behavioral neuroscience, 13:86, 2019.
[36] Soyiba Jawed, Hafeez Ullah Amin, Aamir Saeed Malik, and Ibrahima Faye. Eeg visual and non-visual learner classification using lstm recurrent neural networks. In 2018 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES), pages 467–471. IEEE.
[37] Soyiba Jawed, Hafeez Ullah Amin, Aamir Saeed Malik, and Ibrahima Faye. Differentiating between visual and non-visual learners using eeg power spectrum entropy. In 2018 International Conference on Intelligent and Advanced System (ICIAS), pages 1–4. IEEE.
[38] Khawla Alhasan, Suleiman Aliyu, Liming Chen, and Feng Chen. Ica-based eeg feature analysis and classification of learning styles. In 2019 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech), pages 271–276. IEEE.
[39] MSA Megat Ali, Aisyah Hartini Jahidin, Nooritawati Md Tahir, and Mohd Nasir Taib. Learning style classification via eeg sub-band spectral centroid frequency features. International Journal of Electrical and Computer Engineering, 4:931–938, 2014.
[40] Nazre Abdul Rashid, Mohd Nasir Taib, Sahrim Lias, Norizam Sulaiman, Zunairah Hj Murat, and Ros Shilawani S Abdul Kadir. Learners’ learning style classification related to iq and stress based on eeg. Procedia-Social and Behavioral Sciences, 29:1061–1070, 2011.
[41] Sarawin Thepsatitporn and Chailerd Pichitpornchai. Visual event-related potential studies supporting the validity of vark learning styles’ visual and read/write learners. Advances in Physiology Education, 40(2):206–212, 2016.
[42] Lawal Ibrahim Faruk Dutsinma and P. Temdee. Vark learning style classification using decision tree with physiological signals. Wireless Personal Communications, pages 1–22, 2020.
[43] https://www.youtube.com/watch?v=ifrH5Z-Ob8o.
[44] https://www.youtube.com/watch?v=ehChauN5J8k.
[45] https://www.bilibili.com/video/BV19Z4y1P7Cq/?from=seopage.
[46] Marius Klug and Klaus Gramann. Identifying key factors for improving ica-based decomposition of eeg data in mobile and stationary experiments. European Journal of Neuroscience, 54(12):8406–8420, 2021.
[47] Luca Pion-Tonachini, Ken Kreutz-Delgado, and Scott Makeig. Iclabel: An automated electroencephalographic independent component classifier, dataset, and website. NeuroImage, 198:181–197, 2019.
[48] Cyril R Pernet, Ramon Martinez-Cancino, Dung Truong, Scott Makeig, and Arnaud Delorme. From bids-formatted eeg data to sensor-space group results: A fully reproducible workflow with eeglab and limo eeg. Frontiers in Neuroscience, 14:610388, 2021.
[49] Joshua S Richman and J Randall Moorman. Physiological time-series analysis using approximate entropy and sample entropy. American Journal of Physiology-Heart and Circulatory Physiology, 278(6):H2039–H2049, 2000.
[50] Pega Zarjam, Julien Epps, and Nigel H Lovell. Characterizing mental load in an arithmetic task using entropy-based features. In 2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA), pages 199–204. IEEE, 2012.
[51] Pega Zarjam, Julien Epps, and Fang Chen. Spectral eeg features for evaluating cognitive load. In 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pages 3841–3844. IEEE, 2011.
[52] Reza Khosrowabadi, Chai Quek, Kai Keng Ang, Sau Wai Tung, and Michel Heijnen. A brain-computer interface for classifying eeg correlates of chronic mental stress. In The 2011 international joint conference on neural networks, pages 757–762. IEEE, 2011.
[53] Abdullah Basuki Rahmat and Keiji Iramina. Classification of multiclass eeg signal related to mental task using higuchi fractal dimension and 10-statistic parameters-support vector machine. In TENCON 2015-2015 IEEE Region 10 Conference, pages 1–6. IEEE, 2015.
[54] Pega Zarjam, Julien Epps, and Fang Chen. Evaluation of working memory load using eeg signals. In Proc. APSIPA Annual Summit and Conference, pages 715–719, 2010.
[55] Chai Tong Yuen, Woo San San, Tan Ching Seong, and Mohamed Rizon Classification of human emotions from eeg signals using statistical features and neural network. International Journal of Integrated Engineering, 1(3), 2009.
[56] Rahul Desai, Pratik Porob, Penjo Rebelo, Damodar Reddy Edla, and Anushree Bablani. Eeg data classification for mental state analysis using wavelet packet transform and gaussian process classifier. Wireless Personal Communications, 115(3):2149–2169, 2020.