Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: fa
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# ParsBERT (v2.0)
|
| 7 |
+
A Transformer-based Model for Persian Language Understanding
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
We reconstructed the vocabulary and fine-tuned the ParsBERT v1.1 on the new Persian corpora in order to provide some functionalities for using ParsBERT in other scopes!
|
| 11 |
+
Please follow the [ParsBERT](https://github.com/hooshvare/parsbert) repo for the latest information about previous and current models.
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
## Persian Text Classification [DigiMag, Persian News]
|
| 15 |
+
|
| 16 |
+
The task target is labeling texts in a supervised manner in both existing datasets `DigiMag` and `Persian News`.
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
### DigiMag
|
| 21 |
+
|
| 22 |
+
A total of 8,515 articles scraped from [Digikala Online Magazine](https://www.digikala.com/mag/). This dataset includes seven different classes.
|
| 23 |
+
|
| 24 |
+
1. Video Games
|
| 25 |
+
2. Shopping Guide
|
| 26 |
+
3. Health Beauty
|
| 27 |
+
4. Science Technology
|
| 28 |
+
5. General
|
| 29 |
+
6. Art Cinema
|
| 30 |
+
7. Books Literature
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
| Label | # |
|
| 34 |
+
|:------------------:|:----:|
|
| 35 |
+
| Video Games | 1967 |
|
| 36 |
+
| Shopping Guide | 125 |
|
| 37 |
+
| Health Beauty | 1610 |
|
| 38 |
+
| Science Technology | 2772 |
|
| 39 |
+
| General | 120 |
|
| 40 |
+
| Art Cinema | 1667 |
|
| 41 |
+
| Books Literature | 254 |
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
**Download**
|
| 45 |
+
You can download the dataset from [here](https://drive.google.com/uc?id=1YgrCYY-Z0h2z0-PfWVfOGt1Tv0JDI-qz)
|
| 46 |
+
|
| 47 |
+
|
| 48 |
+
## Results
|
| 49 |
+
|
| 50 |
+
The following table summarizes the F1 score obtained by ParsBERT as compared to other models and architectures.
|
| 51 |
+
|
| 52 |
+
| Dataset | ParsBERT v2 | ParsBERT v1 | mBERT |
|
| 53 |
+
|:-----------------:|:-----------:|:-----------:|:-----:|
|
| 54 |
+
| Digikala Magazine | 93.65* | 93.59 | 90.72 |
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
## How to use :hugs:
|
| 59 |
+
|
| 60 |
+
| Task | Notebook |
|
| 61 |
+
|---------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
| 62 |
+
| Text Classification | [](https://colab.research.google.com/github/hooshvare/parsbert/blob/master/notebooks/Taaghche_Sentiment_Analysis.ipynb) |
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
### BibTeX entry and citation info
|
| 66 |
+
|
| 67 |
+
Please cite in publications as the following:
|
| 68 |
+
|
| 69 |
+
```bibtex
|
| 70 |
+
@article{ParsBERT,
|
| 71 |
+
title={ParsBERT: Transformer-based Model for Persian Language Understanding},
|
| 72 |
+
author={Mehrdad Farahani, Mohammad Gharachorloo, Marzieh Farahani, Mohammad Manthouri},
|
| 73 |
+
journal={ArXiv},
|
| 74 |
+
year={2020},
|
| 75 |
+
volume={abs/2005.12515}
|
| 76 |
+
}
|
| 77 |
+
```
|
| 78 |
+
|
| 79 |
+
## Questions?
|
| 80 |
+
Post a Github issue on the [ParsBERT Issues](https://github.com/hooshvare/parsbert/issues) repo.
|