Uni-LSDPM: A Unified Online Learning Session Dropout Prediction Model Based on Pre-Training

Chen Rui; Wang Zhanquan

doi:10.7544/issn1000-1239.202220834

Journal of Computer Research and Development > 2024 > 61(2): 441-459. > DOI: 10.7544/issn1000-1239.202220834 CSTR: 32373.14.issn1000-1239.202220834

Chen Rui, Wang Zhanquan. Uni-LSDPM: A Unified Online Learning Session Dropout Prediction Model Based on Pre-Training[J]. Journal of Computer Research and Development, 2024, 61(2): 441-459. DOI: 10.7544/issn1000-1239.202220834

Citation:

PDF (2011 KB)

Uni-LSDPM: A Unified Online Learning Session Dropout Prediction Model Based on Pre-Training

Chen Rui,
Wang Zhanquan^,

College of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237

More Information

Author Bio:
Chen Rui: born in 1997. Master candidate. Her main research interests include intelligent education technology, big data mining, and personality prediction

Wang Zhanquan: born in 1975. PhD, professor. His main research interests include big data mining and intelligent education technology
Received Date: September 27, 2022
Revised Date: April 05, 2023
Available Online: November 13, 2023

Graphical Abstract

Abstract

Abstract

To assist learners in maintaining the continuity of online learning and guiding the implementation of the optimal learning path, the intelligent tutoring system (ITS) needs to detect the tendency of learners to withdraw from learning in time and take corresponding intervention measures at the right time. Therefore, online learning session dropout prediction research is necessary. However, compared with traditional course dropouts, session dropouts occur more frequently and with shorter single study sessions. Session dropout requires the accurate prediction of learning session dropout state based on limited learning behavior feature data. Therefore, the fragmentation of learning behavior and the immediacy and accuracy of prediction results are the challenges and difficulties of learning session dropout prediction tasks. For the session dropout prediction task, we propose a unified online learning session dropout prediction model (Uni-LSDPM). Based on the pre-training and fine-tuning paradigm, Uni-LSDPM uses a multi-layer Transformer structure. In the pre-training stage, a bidirectional attention mechanism is used to learn the representation of sequence of learners’ continuous behavioral interaction features. In the fine-tuning stage, a sequence-to-sequence (Seq2Seq) attentional mechanism is used to learn the sequence combination of the learner’s continuous behavioral interaction features and dropout state. The proposed model is pre-trained and fine-tuned based on the EdNet public dataset, and the best prediction effect is obtained through ablation experiment. Comparative experiments are conducted based on multiple datasets. Experimental results show that Uni-LSDPM outperforms existing models in terms of AUC and ACC, and prove that the model has certain robustness and scalability.
- attention mechanism,
- learning session,
- dropout prediction,
- intelligent tutoring system (ITS),
- pre-training

FullText(HTML)

References (63)

References

[1]	Son N T, Jaafar J, Aziz I A, et al. Meta-heuristic algorithms for learning path recommender at MOOC[J]. IEEE Access, 2021, 9: 59093−59107 doi: 10.1109/ACCESS.2021.3072222
[2]	Wanichsan D, Panjaburee P, Chookaew S. Enhancing knowledge integration from multiple experts to guiding personalized learning paths for testing and diagnostic systems[J]. Computers and Education: Artificial Intelligence, 2021, 2: 100013 doi: 10.1016/j.caeai.2021.100013
[3]	Ramos D B, Ramos I M M, Gasparini I, et al. A new learning path model for e-learning systems[J]. International Journal of Distance Education Technologies, 2021, 19(2): 34−54 doi: 10.4018/IJDET.20210401.oa2
[4]	Meng Lingling, Zhang Wanxue, Chu Yu, et al. LD–LP generation of personalized learning path based on learning diagnosis[J]. IEEE Transactions on Learning Technologies, 2021, 14(1): 122−128 doi: 10.1109/TLT.2021.3058525
[5]	Aldowah H, Al-Samarraie H, Alzahrani A I, et al. Factors affecting student dropout in MOOCs: A cause and effect decision‐making model[J]. Journal of Computing in Higher Education, 2020, 32(2): 429−454 doi: 10.1007/s12528-019-09241-y
[6]	Zhang Jingjing, Gao Ming, Zhang Jiang. The learning behaviours of dropouts in MOOCs: A collective attention network perspective[J]. Computers & Education, 2021, 167: 104189
[7]	Eriksson T, Adawi T, Stöhr C. “Time is the bottleneck”: A qualitative study exploring why learners drop out of MOOCs[J]. Journal of Computing in Higher Education, 2017, 29(1): 133−146 doi: 10.1007/s12528-016-9127-8
[8]	Choi H J, Park J H. Testing a path-analytic model of adult dropout in online degree programs[J]. Computers & Education, 2018, 116: 130−138
[9]	Borrella I, Caballero-Caballero S, Ponce-Cueto E. Taking action to reduce dropout in MOOCs: Tested interventions[J]. Computers & Education, 2022, 179: 104412
[10]	Halfaker A, Keyes O, Kluver D, et al. User session identification based on strong regularities in inter-activity time[C] //Proc of the 24th Int Conf on World Wide Web. New York: ACM, 2015: 410−418
[11]	Prenkaj B, Stilo G, Madeddu L. Challenges and solutions to the student dropout prediction problem in online courses[C] //Proc of the 29th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2020: 3513−3514
[12]	Prenkaj B, Velardi P, Stilo G, et al. A survey of machine learning approaches for student dropout prediction in online courses[J]. ACM Computing Surveys, 2020, 53(3): 1−34
[13]	Dalipi F, Imran A S, Kastrati Z. MOOC dropout prediction using machine learning techniques: Review and research challenges[C]// Proc of the 2018 IEEE Global Engineering Education Conf (EDUCON). Piscataway, NJ: IEEE, 2018: 1007−1014
[14]	Alshabandar R, Hussain A, Keight R, et al. Students performance prediction in online courses using machine learning algorithms[C/OL]// Proc of the 2020 Int Joint Conf on Neural Networks (IJCNN). Piscataway, NJ: IEEE, 2020[2022-08-20].https://ieeexplore.ieee.org/document/9207196
[15]	De Oliveira M M, Barwaldt R, Pias M R, et al. Understanding the student dropout in distance learning[C/OL]// Proc of the 2019 IEEE Frontiers in Education Conf (FIE). Piscataway, NJ: IEEE, 2019[2022-08-20].https://ieeexplore.ieee.org/document/9028433
[16]	Qiu Lin, Liu Yanshen, Hu Quan, et al. Student dropout prediction in massive open online courses by convolutional neural networks[J]. Soft Computing, 2019, 23(20): 10287−10301 doi: 10.1007/s00500-018-3581-3
[17]	Olive D M, Huynh D Q, Reynolds M, et al. A quest for a one-size-fits-all neural network: Early prediction of students at risk in online courses[J]. IEEE Transactions on Learning Technologies, 2019, 12(2): 171−183 doi: 10.1109/TLT.2019.2911068
[18]	Xing Wanli, Du Dongping. Dropout prediction in MOOCs: Using deep learning for personalized intervention[J]. Journal of Educational Computing Research, 2019, 57(3): 547−570 doi: 10.1177/0735633118757015
[19]	Lee Y, Shin D, Loh H B, et al. Deep attentive study session dropout prediction in mobile learning environment[C] //Proc of the 12th Int Conf on Computer Supported Education. Portugal: SciTePress, 2020: 26−35
[20]	Jacob D, Chang Mingwei, Kenton L, et al. Bert: Pre-training of deep bidirectional transformers for language understanding [C] //Proc of the 2019 Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg, PA: ACL, 2019: 4171−4186
[21]	Dong Li, Yang Nan, Wang Wenhui, et al. Unified language model pre-training for natural language understanding and generation[C] //Proc of the 33rd Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2019: 13063–13075
[22]	Dass S, Gary K, Cunningham J. Predicting student dropout in self-paced MOOC course using random forest model[J]. Information, 2021, 12(11): 476 doi: 10.3390/info12110476
[23]	Hong Bowei, Wei Zhiqiang, Yang Yongquan. A two-layer cascading method for dropout prediction in MOOC[J]. Mechatronic Systems and Control, 2019, 47(2): 91−97
[24]	Coussement K, Phan M, De Caigny A, et al. Predicting student dropout in subscription-based online learning environments: The beneficial impact of the logit leaf model[J]. Decision Support Systems, 2020, 135: 113325 doi: 10.1016/j.dss.2020.113325
[25]	Lee J H, Kim M, Kim D, et al. Evaluation of predictive models for early identification of dropout students[J]. Journal of Information Processing Systems, 2021, 17(3): 630−644
[26]	Boudjehem R, Lafifi Y. A new approach to identify dropout learners based on their performance-based behavior[J]. Journal of Universal Computer Science, 2021, 27(10): 1001−1025 doi: 10.3897/jucs.74280
[27]	Alamri A, Sun Z, Cristea A I, et al. MOOC next week dropout prediction: Weekly assessing time and learning patterns[C] //Proc of the 17th Int Conf on Intelligent Tutoring Systems. Berlin: Springer, 2021: 119−130
[28]	Jin Cong. Dropout prediction model in MOOC based on clickstream data and student sample weight[J]. Soft Computing, 2021, 25(14): 8971−8988 doi: 10.1007/s00500-021-05795-1
[29]	Imran A S, Dalipi F, Kastrati Z. Predicting student dropout in a MOOC: An evaluation of a deep neural network model[C] //Proc of the 5th Int Conf on Computing and Artificial Intelligence. New York: ACM, 2019: 190−195
[30]	Wu Dongen, Hao Pengyi, Zheng Yuxiang, et al. Classmates enhanced diversity-self-attention network for dropout prediction in MOOCs[C] //Proc of the 28th Int Conf on Neural Information Processing. Berlin: Springer, 2021: 609−620
[31]	Goel Y, Goyal R. On the effectiveness of self-training in mooc dropout prediction[J]. Open Computer Science, 2020, 10(1): 246−258 doi: 10.1515/comp-2020-0153
[32]	Zhang Yan, Chang Liang, Liu Tieyuan. MOOCs dropout prediction based on hybrid deep neural network[C] //Proc of the 2020 Int Conf on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC). Piscataway, NJ: IEEE, 2020: 197−203
[33]	Wu Nannan, Zhang Lei, Gao Yi, et al. CLMS-Net: Dropout prediction in MOOCs with deep learning[C/OL] //Proc of the 2019 ACM Turing Celebration Conf-China. New York: ACM, 2019[2022-08-20].https://dl.acm.org/doi/10.1145/3321408.3322848
[34]	Feng Wenzheng, Tang Jie, Liu T X. Understanding dropouts in MOOCs[C] //Proc of the 33rd AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2019: 517−524
[35]	Fu Qian, Gao Zhanghao, Zhou Junyi, et al. CLSA: A novel deep learning model for MOOC dropout prediction[J]. Computers & Electrical Engineering, 2021, 94: 107315
[36]	Mubarak A A, Cao Han, Hezam I M. Deep analytic model for student dropout prediction in massive open online courses[J]. Computers & Electrical Engineering, 2021, 93: 107271
[37]	Wang Lin, Yu Zhengfei, Wang Mengru, et al. MOOC dropout prediction based on dynamic embedding representation learning[C/OL] //Proc of the 5th Int Conf on Computer Science and Application Engineering. New York: ACM, 2021[2022-08-20].https://dl.acm.org/doi/10.1145/3487075.3487141
[38]	Nitta I, Ishizaki R, Shingu M, et al. Graph-based massive open online course (MOOC) dropout prediction using clickstream data in virtual learning environment[C] //Proc of the 16th Int Conf on Computer Science & Education (ICCSE). Piscataway, NJ: IEEE, 2021: 48−52
[39]	Ramírez Luelmo S I, El Mawas N, Heutte J. Machine learning techniques for knowledge tracing: A systematic literature review[C] //Proc of the 13th Int Conf on Computer Supported Education. Portugal: SciTePress, 2021: 60−70
[40]	De Oliveira T N, Bernardini F, Viterbo J. An overview on the use of educational data mining for constructing recommendation systems to mitigate retention in higher education[C/OL]// Proc of the 2021 IEEE Frontiers in Education Conf (FIE). Piscataway, NJ: IEEE, 2021[2022-08-20].https://ieeexplore.ieee.org/document/9637207
[41]	Song Xiangyu, Li Jianxin, Cai Taotao, et al. A survey on deep learning based knowledge tracing[J]. Knowledge-Based Systems, 2022, 258: 110036 doi: 10.1016/j.knosys.2022.110036
[42]	Shen Shuanghong, Liu Qi, Chen Enhong, et al. Convolutional knowledge tracing: Modeling individualization in student learning process[C] //Proc of the 43rd Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2020: 1857−1860
[43]	Sun Xia, Zhao Xu, Li Bo, et al. Dynamic key-value memory networks with rich features for knowledge tracing[J]. IEEE Transactions on Cybernetics, 2021, 52(8): 8239−8245
[44]	Ghosh A, Heffernan N, Lan A S. Context-aware attentive knowledge tracing[C]//Proc of the 26th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining. New York: ACM, 2020: 2330−2339
[45]	Liu Qi, Huang Zhenya, Yin Yu, et al. Ekt: Exercise-aware knowledge tracing for student performance prediction[J]. IEEE Transactions on Knowledge and Data Engineering, 2019, 33(1): 100−115
[46]	Pandey S, Karypis G. A self-attentive model for knowledge tracing[J]. arXiv preprint, arXiv: 1907. 06837, 2019
[47]	Zhang Wei, Qu Kaiyuan, Han Yahui, et al. A novel knowledge tracing model based on collaborative multi-head attention[C]// Proc of the 6th Int Conf on Innovation in Artificial Intelligence (ICIAI). New York: ACM, 2022: 210−215
[48]	Tan Weicong, Jin Yuan, Liu Ming, et al. BiDKT: Deep knowledge tracing with BERT[C] //Proc of the 2021 Int Conf on Ad Hoc Networks, Proc of the 2021 Int Conf on Testbeds and Research Infrastructures. Berlin: Springer, 2022: 260−278
[49]	Ma Yuling, Han Peng, Qiao Huiyan, et al. SPAKT: A self-supervised pre-training method for knowledge tracing[J]. IEEE Access, 2022, 10: 72145−72154 doi: 10.1109/ACCESS.2022.3187987
[50]	Schrumpf J, Weber F, Thelen T. A neural natural language processing system for educational resource knowledge domain classification[C/OL]// Proc of the 19th Fachtagung Bildungstechnologien (DELFI). 2021[2022-08-20].https://dl.gi.de/handle/20.500.12116/37023
[51]	Schrumpf J, Thelen T. Re-thinking Transformer based educational resource recommendation engines for higher education[C/OL]// Proc of the 20th Fachtagung Bildungstechnologien (DELFI). 2022[2022-08-20].https://dl.gi.de/handle/20.500.12116/38852
[52]	Lebreton P, Yamagishi K. Study on user quitting in the puffer live TV video streaming service[C]// Proc of the 13th Int Conf on Quality of Multimedia Experience (QoMEX). Piscataway, NJ: IEEE, 2021: 19−24
[53]	Hatt T, Feuerriegel S. Early detection of user exits from clickstream data: A Markov modulated marked point process model[C]//Proc of the Web Conf. New York: ACM, 2020: 1671−1681
[54]	Karumbaiah S, Baker R S, Shute V. Predicting quitting in students playing a learning game[C/OL]//Proc of the 11th Int Conf on Educational Data Mining (EDM). 2018[2022-08-20].https://educationaldatamining.org/files/ conferences/EDM2018/papers/EDM2018_paper_39.pdf
[55]	Deeva G, Smedt J D, Koninck P D, et al. Dropout prediction in MOOCs: A comparison between process and sequence mining[C] //Proc of the 2017 Int Conf on Business Process Management. Berlin: Springer, 2017: 243−255
[56]	Liu Kai, Tatinati S, Khong A W H. A weighted feature extraction technique based on temporal accumulation of learner behavior features for early prediction of dropouts[C]// Proc of the 2020 IEEE Int Conf on Teaching, Assessment, and Learning for Engineering (TALE). Piscataway, NJ: IEEE, 2020: 295−302
[57]	Rzepka N, Simbeck K, Müller H G, et al. Keep It Up: In-session dropout prediction to support blended classroom scenarios[C]// Proc of the 14th Int Conf on Computer Supported Education. Portugal: SciTePress, 2022: 131−138
[58]	He Yanbai, Chen Rui, Li Xinya, et al. Online at-risk student identification using RNN-GRU joint neural networks[J]. Information, 2020, 11(10): 474 doi: 10.3390/info11100474
[59]	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C] // Proc of the 30th Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2017: 5998−6008
[60]	Yin Shengjun, Lei Leqi, Wang Hongzhi, et al. Power of attention in MOOC dropout prediction[J]. IEEE Access, 2020, 8: 202993−203002 doi: 10.1109/ACCESS.2020.3035687
[61]	Pulikottil S C, Gupta M. Onet—A temporal meta embedding network for mooc dropout prediction[C] //Proc of the 2020 IEEE Int Conf on Big Data (Big Data). Piscataway, NJ: IEEE, 2020: 5209−5217
[62]	Choi Y, Lee Y, Shin D, et al. Ednet: A large-scale hierarchical dataset in education[C] //Proc of the 21st Int Conf on Artificial Intelligence in Education. Berlin: Springer, 2020: 69−73
[63]	Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: Synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321−357 doi: 10.1613/jair.953

[1]	Fu Nan, Ni Weiwei, Jiang Zepeng, Hou Lihe, Zhang Dongyue, Zhang Ruyu. Directed Graph Clustering Algorithm with Edge Local Differential Privacy[J]. Journal of Computer Research and Development, 2025, 62(1): 256-268. DOI: 10.7544/issn1000-1239.202330193
[2]	Deng Xinguo, Zhang Xinhong, Chen Jiarui, Liu Qinghai, Chen Chuandong. A Weighted Directed Graph-Based Algorithm for Group Routing in Printed Circuit Boards[J]. Journal of Computer Research and Development. DOI: 10.7544/issn1000-1239.202440069
[3]	Shang Jing, Wu Zhihui, Xiao Zhiwen, Zhang Yifei. Graph4Cache: A Graph Neural Network Model for Cache Prefetching[J]. Journal of Computer Research and Development, 2024, 61(8): 1945-1956. DOI: 10.7544/issn1000-1239.202440190
[4]	Zhang Tianming, Xu Yiheng, Cai Xinwei, Fan Jing. A Shortest Path Query Method over Temporal Graphs[J]. Journal of Computer Research and Development, 2022, 59(2): 362-375. DOI: 10.7544/issn1000-1239.20210893
[5]	Zhang Xin, Li Xiaoguang. Spanner Algorithm for Directed Graph Stream[J]. Journal of Computer Research and Development, 2019, 56(3): 655-665. DOI: 10.7544/issn1000-1239.2019.20170680
[6]	Wang Juanjuan, Qiao Ying, Wang Hongan. Graph-Based Auto-Driving Reasoning Task Scheduling[J]. Journal of Computer Research and Development, 2017, 54(8): 1693-1702. DOI: 10.7544/issn1000-1239.2017.20170212
[7]	He Wei, Zhao Ruilian, and Zhu Qunxiong. Call-Graph-Based Interclass MM Path Generation[J]. Journal of Computer Research and Development, 2013, 50(2): 332-343.
[8]	Xu Shifeng, Gao Jun, Yang Dongqing, and Wang Tengjiao. Pass-Count-Based Path Query on Big Graph Datasets[J]. Journal of Computer Research and Development, 2010, 47(1): 96-103.
[9]	Yao Guohui, Zhu Daming, and Ma Shaohan. Approximating the Directed Minimum Degree Spanning Tree of Directed Acyclic Graph[J]. Journal of Computer Research and Development, 2009, 46(6): 1052-1057.
[10]	Wen Yanzhi, Lian Ruiqi, Wu Chengyong, Feng Xiaobing, and Zhang Zhaoqing. A Micro-Scheduling Method on Directed Cyclic Graph[J]. Journal of Computer Research and Development, 2005, 42(3).