Transformers
Safetensors
English
t5
text2text-generation
aspect-based-sentiment-analysis
absa
natural-language-processing
masked-absa
text-generation-inference
Instructions to use tweetpie/masked-absa-rest15 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tweetpie/masked-absa-rest15 with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("tweetpie/masked-absa-rest15") model = AutoModelForSeq2SeqLM.from_pretrained("tweetpie/masked-absa-rest15", device_map="auto") - Notebooks
- Google Colab
- Kaggle
metadata
license: gpl-3.0
language:
- en
widget:
- text: I liked [MASK] but the service was awful.
- text: I liked sushi but the [MASK] was awful.
tags:
- aspect-based-sentiment-analysis
- absa
- natural-language-processing
- masked-absa
Masking The Bias : From Echo Chambers to Large Scale Aspect-Based Sentiment Analysis
This repository addresses the challenges of scarcity and high costs associated with manual annotation in Aspect-Based Sentiment Analysis (ABSA) by utilizing weak supervision methods. We have demonstrated the superiority of our model using SemEval datasets and real-world tweet data.
For more details please visit github.com/tweetpie/masked-absa