Update README.md
Browse files
README.md
CHANGED
@@ -10,5 +10,19 @@ tags:
|
|
10 |
- Politics
|
11 |
- Twitter
|
12 |
---
|
13 |
-
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
- Politics
|
11 |
- Twitter
|
12 |
---
|
13 |
+
# Model Description
|
14 |
+
This model adapts [Moritz Laurer's](https://huggingface.co/MoritzLaurer/deberta-v3-large-zeroshot-v1.1-all-33 ) zero shot model for political texts.
|
15 |
+
It is currently trained for zero-shot classification of stances towards political groups and people, although it should also preform well for topic and issue stance classification.
|
16 |
+
Further capabilities will be added and benchmarked as more training data is developed.
|
17 |
+
|
18 |
+
# Training Data
|
19 |
+
The model was trained using the [PoliStance Affect](https://huggingface.co/datasets/mlburnham/PoliStance_Affect) dataset.
|
20 |
+
The data contains ~27,000 political texts about U.S. politicians and political groups that have been triple coded for stance.
|
21 |
+
The test set contains documents about six politicians that were not included in the training set in order to evaluate zero-shot classification performance.
|
22 |
+
|
23 |
+
# Evaluation
|
24 |
+
Results below are performance on the PoliStance Affect test set.
|
25 |
+
|
26 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/64d0341901931c60161f2a06/NLJtILuPLKtxN0bJJwD0C.png" width="750" height="500" />
|
27 |
+
|
28 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/64d0341901931c60161f2a06/4tOqiINS6BWItRklrqkgY.png" width="750" height="500" />
|