---
license: mit
---



This model is primiarly used to automatically grade the English essay partically written by L2. The training dataset is the English Language Learner Insight, Proficiency and Skills Evaluation (ELLIPSE) Corpus which is a freely available corpus of ~6,500 ELL writing samples that have been scored for overall holistic language proficiency as well as analytic proficiency scores related to cohesion, syntax, vocabulary, phraseology, grammar, and conventions. 


The performance of this model in the test dataset (about 980 English essays) is as follows:  'accuracy':, `f1 score':


The output from this model after an essey is input would be like:  the six scores on cohesion, syntax, vocabulary, phraseology, grammar, and conventions. Each score ranges from 1-5, and the higher scores indicates higher proficiency in this essay. The six dimension could measure the quality of the input essay from different perspectives. This model could be useful for EFL teachers and researchers, as well as useful to English L2 and parents to make self-evaluations for their composition levels.


Run the following code to test the model, or you paste your essay into the API bar

```


```