Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# DziriBERT — Algerian Darija Misinformation Detection
|
| 2 |
|
| 3 |
**DziriBERT** is a fine-tuned **XLM-RoBERTa-large** model for detecting misinformation in **Algerian Darija** text from social media and news.
|
|
@@ -46,7 +63,6 @@
|
|
| 46 |
- Strong performance on Fake, Real, and Non-new classes
|
| 47 |
- Handles Darija, Arabic, and French code-switching well
|
| 48 |
|
| 49 |
-
|
| 50 |
**Limitations**
|
| 51 |
- Low performance on Satire due to limited samples
|
| 52 |
- Misleading class remains challenging
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ar
|
| 4 |
+
- fr
|
| 5 |
+
license: mit
|
| 6 |
+
pipeline_tag: text-classification
|
| 7 |
+
tags:
|
| 8 |
+
- misinformation-detection
|
| 9 |
+
- fake-news
|
| 10 |
+
- text-classification
|
| 11 |
+
- algerian-darija
|
| 12 |
+
- arabic
|
| 13 |
+
- xlm-roberta
|
| 14 |
+
model_name: DziriBERT
|
| 15 |
+
base_model: xlm-roberta-large
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
# DziriBERT — Algerian Darija Misinformation Detection
|
| 19 |
|
| 20 |
**DziriBERT** is a fine-tuned **XLM-RoBERTa-large** model for detecting misinformation in **Algerian Darija** text from social media and news.
|
|
|
|
| 63 |
- Strong performance on Fake, Real, and Non-new classes
|
| 64 |
- Handles Darija, Arabic, and French code-switching well
|
| 65 |
|
|
|
|
| 66 |
**Limitations**
|
| 67 |
- Low performance on Satire due to limited samples
|
| 68 |
- Misleading class remains challenging
|