Revealing at-risk learning patterns and corresponding self-regulated strategies via LSTM encoder and time-series clustering

Purpose This study aims to propose a learning pattern analysis method which can improve a predictive model’s performance, as well as discover hidden insights into micro-level learning pattern. Analyzing student’s learning patterns can help instructors understand how their course design or activities shape learning behaviors; depict students’ beliefs about learning and their motivation; and predict learning performance by analyzing individual students’ learning patterns. Although time-series analysis is one of the most feasible predictive methods for learning pattern analysis, literature-indicated current approaches cannot provide holistic insights about learning patterns for personalized intervention. This study identified at-risk students by micro-level learning pattern analysis and detected pattern types, especially at-risk patterns that existed in the case study. The connections among students’ learning patterns, corresponding self-regulated learning (SRL) strategies and learning performance were finally revealed. Design/methodology/approach The method used long short-term memory (LSTM)-encoder to process micro-level behavioral patterns for feature extraction and compression, thus the students’ behavior pattern information were saved into encoded series. The encoded time-series data were then used for pattern analysis and performance prediction. Time series clustering were performed to interpret the unique strength of proposed method. Findings Successful students showed consistent participation levels and balanced behavioral frequency distributions. The successful students also adjusted learning behaviors to meet with course requirements accordingly. The three at-risk patten types showed the low-engagement (R1) the low-interaction (R2) and the non-persistent characteristics (R3). Successful students showed more complete SRL strategies than failed students. Political Science had higher at-risk chances in all three at-risk types. Computer Science, Earth Science and Economics showed higher chances of having R3 students. Research limitations/implications The study identified multiple learning patterns which can lead to the at-risk situation. However, more studies are needed to validate whether the same at-risk types can be found in other educational settings. In addition, this case study found the distributions of at-risk types were vary in different subjects. The relationship between subjects and at-risk types is worth further investigation. Originality/value This study found the proposed method can effectively extract micro-level behavioral information to generate better prediction outcomes and depict student’s SRL learning strategies in online learning. The authors confirm that the research in their work is original, and that all the data given in the paper are real and authentic. The study has not been submitted to peer review and not has been accepted for publishing in another journal.

[1]  Hiroaki Ogata,et al.  Applying Learning Analytics for the Early Prediction of Students' Academic Performance in Blended Learning , 2018, J. Educ. Technol. Soc..

[2]  Elizabeth Stokoe,et al.  Constructing Topicality in University Students' Small-group Discussion: A Conversation Analytic Approach , 2000 .

[3]  Nicolás Morales,et al.  Mining theory-based patterns from Big data: Identifying self-regulated learning strategies in Massive Open Online Courses , 2018, Comput. Hum. Behav..

[4]  J. Broadbent,et al.  Self-regulated learning strategies & academic achievement in online higher education learning environments: A systematic review , 2015, Internet High. Educ..

[5]  Jürgen Schmidhuber,et al.  Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction , 2011, ICANN.

[6]  Olga Viberg,et al.  Self-regulated learning and learning analytics in online learning environments: a review of empirical research , 2020, LAK.

[7]  P. Reimann,et al.  Process mining techniques for analysing patterns and strategies in students’ self-regulated learning , 2013, Metacognition and Learning.

[8]  Ozren Gamulin,et al.  Using Fourier coefficients in time series analysis for student performance prediction in blended learning environments , 2016, Expert Syst. J. Knowl. Eng..

[9]  Hongxun Yao,et al.  Auto-encoder based dimensionality reduction , 2016, Neurocomputing.

[10]  Richard A. Levine,et al.  Using a Latent Class Forest to Identify At-Risk Students in Higher Education , 2019 .

[11]  Cristina Conati,et al.  Applying a Framework for Student Modeling in Exploratory Learning Environments: Comparing Data Representation Granularity to Handle Environment Complexity , 2017, International Journal of Artificial Intelligence in Education.

[12]  Xu Du,et al.  An Integrated Framework Based on Latent Variational Autoencoder for Providing Early Warning of At-Risk Students , 2020, IEEE Access.

[13]  Chien Chou,et al.  Applying Learning Analytics to Explore the Effects of Motivation on Online Students' Reading Behavioral Patterns , 2018 .

[14]  Yunpeng Wang,et al.  Long short-term memory neural network for traffic speed prediction using remote microwave sensor data , 2015 .

[15]  Rafi Nachmias,et al.  Types of Participant Behavior in a Massive Open Online Course , 2017 .

[16]  M. Boekaerts.  SELF-REGULATED LEARNING: A NEW CONCEPT EMBRACED BY RESEARCHERS, POLICY MAKERS, EDUCATORS, TEACHERS, AND STUDENTS , 1997 .

[17]  Shuicheng Yan,et al.  Robust LSTM-Autoencoders for Face De-Occlusion in the Wild , 2016, IEEE Transactions on Image Processing.

[18]  Ke Wang,et al.  Recovering loss to followup information using denoising autoencoders , 2017, 2017 IEEE International Conference on Big Data (Big Data).

[19]  P. Pintrich.  A Conceptual Framework for Assessing Motivation and Self-Regulated Learning in College Students , 2004 .

[20]  Xu Du,et al.  Two-Stage Predictive Modeling for Identifying At-Risk Students , 2018, ICITL.

[21]  Hiroshi Kato,et al.  Procrastination and other learning behavioral types in e-learning and their relationship with learning outcomes , 2015 .

[22]  Vincent Donche,et al.  A Learning Patterns Perspective on Student Learning in Higher Education: State of the Art and Moving Forward , 2017, Educational Psychology Review.

[23]  Yaohang Li,et al.  Identifying At-Risk Students for Early Interventions—A Time-Series Clustering Approach , 2017, IEEE Transactions on Emerging Topics in Computing.

[24]  B. Zimmerman,et al.  Self-regulation: Where metacognition and motivation intersect. , 2009 .

[25]  Michael Derntl,et al.  A Pattern Approach to Person-Centered e-Learning Based on Theory-Guided Action Research , 2004 .

[26]  Qun Jin,et al.  Goal-Driven Process Navigation for Individualized Learning Activities in Ubiquitous Networking and IoT Environments , 2012, J. Univers. Comput. Sci..

[27]  Francesco Piazza,et al.  Unsupervised electric motor fault detection by using deep autoencoders , 2019, IEEE/CAA Journal of Automatica Sinica.

[28]  Jihong Li,et al.  Behavioral patterns of knowledge construction in online cooperative translation activities , 2018, Internet High. Educ..

[29]  Anouschka van Leeuwen,et al.  The role of temporal patterns in students' behavior for predicting course performance: A comparison of two blended learning courses , 2019, Br. J. Educ. Technol..

[30]  M. Lepper,et al.  Intrinsic and Extrinsic Motivational Orientations in the Classroom: Age Differences and Academic Correlates , 2005 .

[31]  Jaclyn Broadbent,et al.  Comparing online and blended learner's self-regulated learning strategies and academic performance , 2017, Internet High. Educ..

[32]  Dirk Ifenthaler,et al.  Cognitive, metacognitive and motivational perspectives on preflection in self-regulated online learning , 2014, Comput. Hum. Behav..

[33]  B. Shelton,et al.  Is learning anytime, anywhere a good strategy for success? Identifying successful spatial-temporal patterns of on-the-job and full-time students , 2019, Information Discovery and Delivery.

[34]  Philip H. Winne,et al.  How Software Technologies Can Improve Research on Learning and Bolster School Reform , 2006 .

[35]  Ernesto Panadero,et al.  A Review of Self-regulated Learning: Six Models and Four Directions for Research , 2017, Front. Psychol..

[36]  John Cristian Borges Gamboa,et al.  Deep Learning for Time-Series Analysis , 2017, ArXiv.

[37]  Hui Liu.  An Analysis on Blended Learning Pattern Based on Blackboard Network Platform: A Case Study on the Course of Recruitment and Employment Management , 2016, iJET.

[38]  Mung Chiang,et al.  Behavior-Based Grade Prediction for MOOCs Via Time Series Neural Networks , 2017, IEEE Journal of Selected Topics in Signal Processing.

[39]  Lorena Blasco-Arcas,et al.  Using clickers in class. The role of interactivity, active collaborative learning and engagement in learning performance , 2013, Comput. Educ..

[40]  W. E. Wecker,et al.  Predicting the Turning Points of a Time Series , 1979 .

[41]  Wu Zhang,et al.  Student Engagement Predictions in an e-Learning System and Their Impact on Student Course Assessment Scores , 2018, Comput. Intell. Neurosci..

[42]  Kun Li,et al.  MOOC learners' demographics, self-regulated learning strategy, perceived learning and satisfaction: A structural equation modeling approach , 2019, Comput. Educ..

[43]  Ke Zhang,et al.  Revealing Online Learning Behaviors and Activity Patterns and Making Predictions with Data Mining Techniques in Online Teaching , 2008 .

[44]  Yan Zhang,et al.  When and who at risk? Call back at these critical points , 2017, EDM.

[45]  Shengquan Yu,et al.  Effects of cooperative translation on Chinese EFL student levels of interest and self-efficacy in specialized English translation , 2016 .

[46]  Stephanie D. Teasley,et al.  A time series interaction analysis method for building predictive models of learners using log data , 2015, LAK.

[47]  Yinhai Wang,et al.  Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Forecasting Network-wide Traffic State with Missing Values , 2020, Transportation Research Part C: Emerging Technologies.