marianaossilva
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,83 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
language:
|
4 |
+
- pt
|
5 |
+
metrics:
|
6 |
+
- name: Precision
|
7 |
+
type: Precision
|
8 |
+
value: 0.783
|
9 |
+
- name: Recall
|
10 |
+
type: Recall
|
11 |
+
value: 0.774
|
12 |
+
- name: F1-Score
|
13 |
+
type: F1-Score
|
14 |
+
value: 0.779
|
15 |
+
library_name: transformers
|
16 |
+
pipeline_tag: token-classification
|
17 |
+
tags:
|
18 |
+
- BERT
|
19 |
+
- CRF
|
20 |
+
- NER
|
21 |
+
- Portuguese
|
22 |
+
- Literature
|
23 |
+
---
|
24 |
+
|
25 |
+
# LitBERT-CRF
|
26 |
+
|
27 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
28 |
+
|
29 |
+
LitBERT-CRF model is a fine-tuned BERT-CRF architecture specifically designed for Named Entity Recognition (NER) in Portuguese-written literature.
|
30 |
+
|
31 |
+
## Model Details
|
32 |
+
|
33 |
+
### Model Description
|
34 |
+
|
35 |
+
LitBERT-CRF leverages a BERT-CRF architecture, initially pre-trained on the brWaC corpus and fine-tuned on the HAREM dataset for enhanced NER performance in Portuguese.
|
36 |
+
It incorporates domain-specific literary data through Masked Language Modeling (MLM), making it well-suited for identifying named entities in literary texts.
|
37 |
+
|
38 |
+
- **Model type:** BERT-CRF for NER
|
39 |
+
- **Language:** Portuguese
|
40 |
+
- **Fine-tuned from model:** BERT-CRF on brWaC and HAREM
|
41 |
+
|
42 |
+
## Evaluation
|
43 |
+
|
44 |
+
### Testing Data, Factors & Metrics
|
45 |
+
|
46 |
+
#### Testing Data
|
47 |
+
|
48 |
+
PPORTAL_ner dataset
|
49 |
+
|
50 |
+
#### Metrics
|
51 |
+
|
52 |
+
- **Precision**: 0.783
|
53 |
+
- **Recall**: 0.774
|
54 |
+
- **F1-score**: 0.779
|
55 |
+
|
56 |
+
## Citation
|
57 |
+
|
58 |
+
**BibTeX:**
|
59 |
+
```
|
60 |
+
@inproceedings{silva-moro-2024-evaluating,
|
61 |
+
title = "Evaluating Pre-training Strategies for Literary Named Entity Recognition in {P}ortuguese",
|
62 |
+
author = "Silva, Mariana O. and
|
63 |
+
Moro, Mirella M.",
|
64 |
+
editor = "Gamallo, Pablo and
|
65 |
+
Claro, Daniela and
|
66 |
+
Teixeira, Ant{\'o}nio and
|
67 |
+
Real, Livy and
|
68 |
+
Garcia, Marcos and
|
69 |
+
Oliveira, Hugo Gon{\c{c}}alo and
|
70 |
+
Amaro, Raquel",
|
71 |
+
booktitle = "Proceedings of the 16th International Conference on Computational Processing of Portuguese - Vol. 1",
|
72 |
+
month = mar,
|
73 |
+
year = "2024",
|
74 |
+
address = "Santiago de Compostela, Galicia/Spain",
|
75 |
+
publisher = "Association for Computational Lingustics",
|
76 |
+
url = "https://aclanthology.org/2024.propor-1.39",
|
77 |
+
pages = "384--393",
|
78 |
+
}
|
79 |
+
```
|
80 |
+
|
81 |
+
**APA:**
|
82 |
+
|
83 |
+
Mariana O. Silva and Mirella M. Moro. 2024. Evaluating Pre-training Strategies for Literary Named Entity Recognition in Portuguese. In Proceedings of the 16th International Conference on Computational Processing of Portuguese - Vol. 1, pages 384–393, Santiago de Compostela, Galicia/Spain. Association for Computational Lingustics.
|