---
license: mit
language:
- en
tags:
- education
- learning analytics
- educational data mining
---

# Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

This is the EduBERT model used in the [EduBERT: Pretrained Deep Language Models for Learning Analytics](https://arxiv.org/abs/1912.00690) from LAK20. It is a fine-tuned version of DistilBERT on educational data.

## Model Description

We originally trained this model to support Learning Analytics task, showing it performed well on well-known educational text classification task.

## Bias, Risks, and Limitations

The model is provided as-is, and trained on the data described in the paper. Learning Analytics is a complex field, and decisions should not be taken fully automatically by models. This model should be used for analysis and to inform only.

## Citation

**BibTeX:**

```
@inproceedings{clavié2019edubert,
      title={EduBERT: Pretrained Deep Language Models for Learning Analytics}, 
      author={Benjamin Clavié and Kobi Gal},
      year={2020},
      booktitle={Companion Proceedings of the The 10th international Learning Analytics & Knowledge (LAK20)}
}
```