Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction
Facing the problem of massive unlabeled data and limited labeled samples, semi-supervised learning is favored, especially co-training. Standard co-training requires sufficiently redundant and conditionally independent dual views; however, in fact, few dual views exist that satisfy this condition. To...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Axioms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1680/11/5/223 |
| _version_ | 1797501606667747328 |
|---|---|
| author | Huanhuan Wang Libo Xu Zhenrui Huang Jiagong Wang |
| author_facet | Huanhuan Wang Libo Xu Zhenrui Huang Jiagong Wang |
| author_sort | Huanhuan Wang |
| collection | DOAJ |
| description | Facing the problem of massive unlabeled data and limited labeled samples, semi-supervised learning is favored, especially co-training. Standard co-training requires sufficiently redundant and conditionally independent dual views; however, in fact, few dual views exist that satisfy this condition. To solve this problem, we propose a co-training method based on semi-decoupling features, that is, semi-decoupling features based on a known single view and then constructing independent and redundant dual views: (1) take a small number of important features as shared features of the dual views according to the importance of the features; (2) separate the remaining features one by one or in small batches according to the correlation between the features to make “divergent” features of the dual views; (3) combine the shared features and the “divergent” features to construct dual views. In this paper, the experimental dataset was from the edX dataset jointly released by Harvard University and MIT; the evaluation metrics adopted <i>F</i>1, <i>Precision</i>, and <i>Recall</i>. The analysis methods included three experiments: multiple models, iterations, and hyperparameters. The experimental results show that the effect of this model on MOOC learner behavior prediction was better than the other models, and the best prediction result was obtained in iteration 2. These all verify the effectiveness and superiority of this algorithm and provide a scientific and feasible reference for the development of the future education industry. |
| first_indexed | 2024-03-10T03:20:57Z |
| format | Article |
| id | doaj.art-e3939cbcc1f54d48b66748e06ae1bcfd |
| institution | Directory Open Access Journal |
| issn | 2075-1680 |
| language | English |
| last_indexed | 2024-03-10T03:20:57Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Axioms |
| spelling | doaj.art-e3939cbcc1f54d48b66748e06ae1bcfd2023-11-23T10:04:18ZengMDPI AGAxioms2075-16802022-05-0111522310.3390/axioms11050223Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior PredictionHuanhuan Wang0Libo Xu1Zhenrui Huang2Jiagong Wang3School of Computing and Data Engineering, NingboTech University, Ningbo 315000, ChinaSchool of Computing and Data Engineering, NingboTech University, Ningbo 315000, ChinaSchool of Computing and Data Engineering, NingboTech University, Ningbo 315000, ChinaSchool of Computing and Data Engineering, NingboTech University, Ningbo 315000, ChinaFacing the problem of massive unlabeled data and limited labeled samples, semi-supervised learning is favored, especially co-training. Standard co-training requires sufficiently redundant and conditionally independent dual views; however, in fact, few dual views exist that satisfy this condition. To solve this problem, we propose a co-training method based on semi-decoupling features, that is, semi-decoupling features based on a known single view and then constructing independent and redundant dual views: (1) take a small number of important features as shared features of the dual views according to the importance of the features; (2) separate the remaining features one by one or in small batches according to the correlation between the features to make “divergent” features of the dual views; (3) combine the shared features and the “divergent” features to construct dual views. In this paper, the experimental dataset was from the edX dataset jointly released by Harvard University and MIT; the evaluation metrics adopted <i>F</i>1, <i>Precision</i>, and <i>Recall</i>. The analysis methods included three experiments: multiple models, iterations, and hyperparameters. The experimental results show that the effect of this model on MOOC learner behavior prediction was better than the other models, and the best prediction result was obtained in iteration 2. These all verify the effectiveness and superiority of this algorithm and provide a scientific and feasible reference for the development of the future education industry.https://www.mdpi.com/2075-1680/11/5/223semi-supervisedco-trainingsemi-decouplingfeature importancePearson correlation coefficient |
| spellingShingle | Huanhuan Wang Libo Xu Zhenrui Huang Jiagong Wang Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction Axioms semi-supervised co-training semi-decoupling feature importance Pearson correlation coefficient |
| title | Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction |
| title_full | Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction |
| title_fullStr | Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction |
| title_full_unstemmed | Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction |
| title_short | Co-Training Method Based on Semi-Decoupling Features for MOOC Learner Behavior Prediction |
| title_sort | co training method based on semi decoupling features for mooc learner behavior prediction |
| topic | semi-supervised co-training semi-decoupling feature importance Pearson correlation coefficient |
| url | https://www.mdpi.com/2075-1680/11/5/223 |
| work_keys_str_mv | AT huanhuanwang cotrainingmethodbasedonsemidecouplingfeaturesformooclearnerbehaviorprediction AT liboxu cotrainingmethodbasedonsemidecouplingfeaturesformooclearnerbehaviorprediction AT zhenruihuang cotrainingmethodbasedonsemidecouplingfeaturesformooclearnerbehaviorprediction AT jiagongwang cotrainingmethodbasedonsemidecouplingfeaturesformooclearnerbehaviorprediction |