File size: 1,559 Bytes
4538315 ccf7652 4538315 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 |
---
license: mit
datasets:
- mlburnham/PoliStance_Affect
pipeline_tag: zero-shot-classification
language:
- en
library_name: transformers
tags:
- Politics
- Twitter
---
# **This model is deprecated. Please use the [Political DEBATE models](https://huggingface.co/mlburnham/Political_DEBATE_large_v1.0) for better performance**
# Model Description
This model adapts [Moritz Laurer's](https://huggingface.co/MoritzLaurer/deberta-v3-base-zeroshot-v1.1-all-33) zero shot base model for political texts.
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.
Further capabilities will be added and benchmarked as more training data is developed.
# Training Data
The model was trained using the [PoliStance Affect](https://huggingface.co/datasets/mlburnham/PoliStance_Affect) dataset.
The data contains ~27,000 political texts about U.S. politicians and political groups that have been triple coded for stance.
The test set contains documents about six politicians that were not included in the training set in order to evaluate zero-shot classification performance.
# Evaluation
Results below are performance on the PoliStance Affect test set.
<img src="https://cdn-uploads.huggingface.co/production/uploads/64d0341901931c60161f2a06/NLJtILuPLKtxN0bJJwD0C.png" width="750" height="500" />
<img src="https://cdn-uploads.huggingface.co/production/uploads/64d0341901931c60161f2a06/4tOqiINS6BWItRklrqkgY.png" width="750" height="500" /> |