Cargando…

A survival analysis based volatility and sparsity modeling network for student dropout prediction

Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals in Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing a single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve c...

Descripción completa

Detalles Bibliográficos
Autores principales: Pan, Feng, Huang, Bingyao, Zhang, Chunhong, Zhu, Xinning, Wu, Zhenyu, Zhang, Moyu, Ji, Yang, Ma, Zhanfei, Li, Zhengchen
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9071151/
https://www.ncbi.nlm.nih.gov/pubmed/35512010
http://dx.doi.org/10.1371/journal.pone.0267138
_version_ 1784700789360427008
author Pan, Feng
Huang, Bingyao
Zhang, Chunhong
Zhu, Xinning
Wu, Zhenyu
Zhang, Moyu
Ji, Yang
Ma, Zhanfei
Li, Zhengchen
author_facet Pan, Feng
Huang, Bingyao
Zhang, Chunhong
Zhu, Xinning
Wu, Zhenyu
Zhang, Moyu
Ji, Yang
Ma, Zhanfei
Li, Zhengchen
author_sort Pan, Feng
collection PubMed
description Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals in Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing a single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve continuous and consistent predictions over time. However, the volatility and sparsity of data always weaken the models’ performance. Prevailing solutions rely heavily on data pre-processing independent of predictive models, which are labor-intensive and may contaminate authentic data. This paper proposes a Survival Analysis based Volatility and Sparsity Modeling Network (SAVSNet) to address these issues in an end-to-end deep learning framework. Specifically, SAVSNet smooths the volatile time series by convolution network while preserving the original data information using Long-Short Term Memory Network (LSTM). Furthermore, we propose a Time-Missing-Aware LSTM unit to mitigate the impact of data sparsity by integrating informative missingness patterns into the model. A survival analysis loss function is adopted for parameter estimation, and the model outputs monotonically decreasing survival probabilities. In the experiments, we compare the proposed method with state-of-the-art methods in two real-world MOOC datasets, and the experiment results show the effectiveness of our proposed model.
format Online
Article
Text
id pubmed-9071151
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-90711512022-05-06 A survival analysis based volatility and sparsity modeling network for student dropout prediction Pan, Feng Huang, Bingyao Zhang, Chunhong Zhu, Xinning Wu, Zhenyu Zhang, Moyu Ji, Yang Ma, Zhanfei Li, Zhengchen PLoS One Research Article Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals in Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing a single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve continuous and consistent predictions over time. However, the volatility and sparsity of data always weaken the models’ performance. Prevailing solutions rely heavily on data pre-processing independent of predictive models, which are labor-intensive and may contaminate authentic data. This paper proposes a Survival Analysis based Volatility and Sparsity Modeling Network (SAVSNet) to address these issues in an end-to-end deep learning framework. Specifically, SAVSNet smooths the volatile time series by convolution network while preserving the original data information using Long-Short Term Memory Network (LSTM). Furthermore, we propose a Time-Missing-Aware LSTM unit to mitigate the impact of data sparsity by integrating informative missingness patterns into the model. A survival analysis loss function is adopted for parameter estimation, and the model outputs monotonically decreasing survival probabilities. In the experiments, we compare the proposed method with state-of-the-art methods in two real-world MOOC datasets, and the experiment results show the effectiveness of our proposed model. Public Library of Science 2022-05-05 /pmc/articles/PMC9071151/ /pubmed/35512010 http://dx.doi.org/10.1371/journal.pone.0267138 Text en © 2022 Pan et al https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Pan, Feng
Huang, Bingyao
Zhang, Chunhong
Zhu, Xinning
Wu, Zhenyu
Zhang, Moyu
Ji, Yang
Ma, Zhanfei
Li, Zhengchen
A survival analysis based volatility and sparsity modeling network for student dropout prediction
title A survival analysis based volatility and sparsity modeling network for student dropout prediction
title_full A survival analysis based volatility and sparsity modeling network for student dropout prediction
title_fullStr A survival analysis based volatility and sparsity modeling network for student dropout prediction
title_full_unstemmed A survival analysis based volatility and sparsity modeling network for student dropout prediction
title_short A survival analysis based volatility and sparsity modeling network for student dropout prediction
title_sort survival analysis based volatility and sparsity modeling network for student dropout prediction
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9071151/
https://www.ncbi.nlm.nih.gov/pubmed/35512010
http://dx.doi.org/10.1371/journal.pone.0267138
work_keys_str_mv AT panfeng asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT huangbingyao asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhangchunhong asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhuxinning asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT wuzhenyu asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhangmoyu asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT jiyang asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT mazhanfei asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT lizhengchen asurvivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT panfeng survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT huangbingyao survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhangchunhong survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhuxinning survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT wuzhenyu survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT zhangmoyu survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT jiyang survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT mazhanfei survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction
AT lizhengchen survivalanalysisbasedvolatilityandsparsitymodelingnetworkforstudentdropoutprediction