Research on Integrating Forgetting Behavior into Student Models for Online Learning Systems
Bibliographic record
Abstract
Modelling students accurately is an important task in online learning systems. In online learning systems, student models are usually built by dealing with students’ historical data of answering questions as input, and eventually output to what extent the student has mastered a certain knowledge component. To evaluate a student model, a commonly used method, namely knowledge tracing, is to build multiple student models based on multiple continuous historical data as mentioned above, then predict whether or not a student can answer a question correctly, and finally compare predicted results with true results. However, the behavior of students is complicated and unpredictable, thus makes student modelling and knowledge tracing become very difficult tasks. Based on existing research of knowledge tracing, especially deep knowledge tracing which uses LSTM to model students, this paper proposes a novel method of student modeling. Compared with other state-of-the-art student modeling methods, the most significant feature of our modeling methods is our method can take students' forgetting behavior into consideration. Moreover, our modeling method can appropriately handle situations that one question corresponds to multiple knowledge components. To test the performance of our student model, this paper applies our student model to the knowledge tracing task. Based on our experiments in public datasets, when one question corresponds to multiple knowledge components and the length of students' historical data is greater, our model performs better in terms of all metrics compared to state-of-the-art knowledge tracing methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".