Add tags, restructure info
Browse files
README.md
CHANGED
|
@@ -1,3 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
## ParsBERT: Transformer-based Model for Persian Language Understanding
|
| 2 |
|
| 3 |
ParsBERT is a monolingual language model based on Google’s BERT architecture with the same configurations as BERT-Base.
|
|
@@ -41,45 +46,14 @@ PEYMA dataset includes 7,145 sentences with a total of 302,530 tokens from which
|
|
| 41 |
**Download**
|
| 42 |
You can download the dataset from [here](http://nsurl.org/tasks/task-7-named-entity-recognition-ner-for-farsi/)
|
| 43 |
|
| 44 |
-
---
|
| 45 |
-
|
| 46 |
-
### ARMAN
|
| 47 |
-
|
| 48 |
-
ARMAN dataset holds 7,682 sentences with 250,015 sentences tagged over six different classes.
|
| 49 |
-
|
| 50 |
-
1. Organization
|
| 51 |
-
2. Location
|
| 52 |
-
3. Facility
|
| 53 |
-
4. Event
|
| 54 |
-
5. Product
|
| 55 |
-
6. Person
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
| Label | # |
|
| 59 |
-
|:------------:|:-----:|
|
| 60 |
-
| Organization | 30108 |
|
| 61 |
-
| Location | 12924 |
|
| 62 |
-
| Facility | 4458 |
|
| 63 |
-
| Event | 7557 |
|
| 64 |
-
| Product | 4389 |
|
| 65 |
-
| Person | 15645 |
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
**Download**
|
| 70 |
-
You can download the dataset from [here](https://github.com/HaniehP/PersianNER)
|
| 71 |
-
|
| 72 |
-
|
| 73 |
|
| 74 |
## Results
|
| 75 |
|
| 76 |
The following table summarizes the F1 score obtained by ParsBERT as compared to other models and architectures.
|
| 77 |
|
| 78 |
-
| Dataset
|
| 79 |
-
|
| 80 |
-
|
|
| 81 |
-
| PEYMA | 98.79* | - | 90.59 | - | 84.00 | - |
|
| 82 |
-
| ARMAN | 93.10* | 89.9 | 84.03 | 86.55 | - | 77.45 |
|
| 83 |
|
| 84 |
|
| 85 |
## How to use :hugs:
|
| 1 |
+
---
|
| 2 |
+
language: fa
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
## ParsBERT: Transformer-based Model for Persian Language Understanding
|
| 7 |
|
| 8 |
ParsBERT is a monolingual language model based on Google’s BERT architecture with the same configurations as BERT-Base.
|
| 46 |
**Download**
|
| 47 |
You can download the dataset from [here](http://nsurl.org/tasks/task-7-named-entity-recognition-ner-for-farsi/)
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
## Results
|
| 51 |
|
| 52 |
The following table summarizes the F1 score obtained by ParsBERT as compared to other models and architectures.
|
| 53 |
|
| 54 |
+
| Dataset | ParsBERT | MorphoBERT | Beheshti-NER | LSTM-CRF | Rule-Based CRF | BiLSTM-CRF |
|
| 55 |
+
|---------|----------|------------|--------------|----------|----------------|------------|
|
| 56 |
+
| PEYMA | 98.79* | - | 90.59 | - | 84.00 | - |
|
|
|
|
|
|
|
| 57 |
|
| 58 |
|
| 59 |
## How to use :hugs:
|