bertweet base emotion analysis

Introduction

BERTweet-Base-Emotion-Analysis is a model designed for emotion detection in English text. It utilizes the BERTweet architecture and is trained with the EmoEvent corpus. This model is part of the pysentimiento library, which provides tools for sentiment analysis and social NLP tasks.

Architecture

The model is based on BERTweet, a derivative of the BERT architecture optimized for text from social media platforms such as Twitter. It leverages the capabilities of transformers and is implemented using PyTorch.

Training

The model has been trained using the EmoEvent corpus, a multilingual dataset designed for emotion detection in textual data related to various events. Training utilizes third-party datasets like TASS and SEMEval 2017, with licenses specific to those datasets.

Guide: Running Locally

1. Clone the repository linked to the model: finiteautomata/pysentimiento.
2. Install the necessary Python packages, including PyTorch and Hugging Face's Transformers library.
3. Load the BERTweet-Base-Emotion-Analysis model using the Transformers library.
4. Prepare your text data for analysis and run the model to detect emotions.

To optimize performance, consider using cloud-based GPUs such as AWS EC2, Google Cloud Platform, or Microsoft Azure.

License

The pysentimiento library is open-source and intended for non-commercial use and scientific research only. The model is trained with third-party datasets, and users must comply with the licenses of these datasets:

• TASS Dataset: License
• SEMEval 2017 Dataset

For any work utilizing pysentimiento, please cite the corresponding research papers.