CLMS-Net

Towards Finding the Impact of Deep Learning in Educational Time Series Datasets – A Systematic Literature Review

“…In studies [27,28,39,42], all the authors used the same dataset (KDD cup 2015) to predict student dropout, and the result shows above 85% performance in all. It contains 39 courses and seven kinds of student behavioral information such as Access, video, wiki, discussion, navigate, page_close and problem.…”

“…These multiple parameters allow applying a multi-variate time series approach. Though the dataset is the same, imbalanced data is handled only in [28,39].…”

Active Methodology, Educational Data Mining and Learning Analytics: A Systematic Mapping Study

“…According to the extracted data, 19 articles studied the attribute access to the course, since the number of times the user logs into the platform, date and time are important to check their willingness, availability, and motivation for learning. About the login duration attribute, Liang et al (2016); Mishra and Mishra (2018); Heidrich et al (2018); Tomasevic et al (2020); Chen and Zhang (2017); ; Imran et al (2019); Niu et al (2018); Kostopoulos et al (2015); Wu et al (2019); Kostopoulos et al (2019b); Kostopoulos et al (2019a); Santos et al (2015); Santos et al (2016); Ramos et al (2017); Rabelo et al (2017); Ramos et al (2018) Login duration 7 ; Kang and Wang (2018); Kostopoulos et al (2015); Wu et al (2019); Kostopoulos et al (2019a); Ramos et al (2017); Ramos et al (2018) Web browsers 5 Liang et al (2016); Mishra and Mishra (2018); ; ; Kostopoulos et al (2015) Table 11 List of articles by demographic attributes Work schedule 1 Kostopoulos et al (2018b) 7 articles used this data to verify how long the user remains on the teaching platform, from their entry to the time they leave the environment. Finally, 5 articles cited data from web browsers for accessing the teaching platform as an important record, capable of helping to identify possible flaws in the user's interaction process with the learning environment.…”

“…study whether semi-supervised techniquesSelf-Training, Co-Training, Democratic Co-Training, Tri-Training, RASCO and Rel-RASCO, classified with Naive Bayes and C4.5, can be useful in predicting school dropout in the distance higher education. The results of the experiments presented the Tri-Training method with C4.5 as the best accuracy rate, varying between 53.26% and 75.29% Wu et al (2019). propose a Deep Artificial Neural Network model called CLMS-Net, which combines Convolutional Neural Network, Long Short-Term Memory Network and Support Vector Machine for the automatic extraction of data related to student behavior.…”

“…. (2018b);Brandão et al (2019);Ortigosa et al (2019);;Liang et al (2016);Cobos and Olmos (2019);Isidro et al (2018);;Waheed et al (2020);Heidrich et al (2018);Tomasevic et al (2020);Islam et al (2019); Chen and Zhang (2017);Whitehill et al (2017);;Kang and Wang (2018);Niu et al (2018);Kostopoulos et al (2015);Wu et al (2019); Borrella et al (2019); Kostopoulos et al (2019b); Kostopoulos et al (2018a); Kostopoulos et al (2019a); Santos et al (2015); Silva et al (2015); Santos et al (2016); Ramos et al (2017); Dos Santos and Falcão (2017); Rabelo et al (2017); Queiroga et al (2017); Ramos et al (2018); Queiroga et al (2019) Prediction 25 Kostopoulos et al (2018b); Mishra and Mishra (2018); Cobos and Olmos (2019); De La Peña et al (2018); Isidro et al (2018); Wang and Wang (2019); Waheed et al (2020); Heidrich et al (2018); Tomasevic et al (2020); Whitehill et al (2017); Wang et al (2017); Kang and Wang (2018); Imran et al (2019); Niu et al (2018); Kostopoulos et al (2015); Wu et al (2019); Borrella et al (2019); Kostopoulos et al (2019b); Kostopoulos et al (2019a); Silva et al (2015); Santos et al (2016); Ramos et al (2017); Queiroga et al (2017); Ramos et al (2018); Queiroga et al (2019) Clustering 6 Brandão et al (2019); Cobos and Olmos (2019); Macedo et al (2019); Oeda and Hashimoto (2017); Islam et al (2019); Dos Santos and Falcão (2017) Regression 2 Kostopoulos et al (2019b); Kostopoulos et al (2018a) Summary/ Visualization 1 Islam et al (2019) Not Specified 1 Brito et al (2019) form, enables the achievement of positive results, timely and effective interventions.…”