Approaches based on language models for aspect extraction for sentiment analysis in the Portuguese language

Welcome to the repository of code, data, and notebooks related to the dissertation on "Aspect Extraction in Portuguese Language" and the article "Approaches based on language models for aspect extraction for sentiment analysis in the Portuguese language". This repository contains all the resources needed to understand and replicate the analyses and results described in the dissertation.

📖 Description

The work focuses on the analysis and extraction of aspects in texts written in Portuguese. In natural language processing, aspects refer to a product or service's specific points or features mentioned in reviews or comments.

📁 Repository Structure

• corpus/: Collection of texts used to pre-train the models.
• datasets/: Structured datasets used in the work, which may include raw, pre-processed, and stratified data.
• notebooks/: Jupyter notebooks containing exploratory analyses, experiments, and visualizations related to the research.
• others/: Other resources and auxiliary files that do not fit directly into the previous categories.
• results/: Results of the experiments.
• LICENSE: License under which the resources of this repository are distributed.
• requeriments.txt: List of libraries and dependencies required to run the codes and notebooks.

⚙️ Installation and Usage

1. Clone the Repository:

bash git clone git@github.com:jcfneto/aspect-extraction.git

1. Install Dependencies:

Navigate to the repository directory and install the dependencies using pip:

bash cd aspect-extraction pip install -r requeriments.txt

1. Browse the Notebooks:

Use a Jupyter environment to explore and run the notebooks in the notebooks/ directory.

📜 License

This project is licensed under the MIT License. See the LICENSE file for more details.

📌 Citation

If you use resources from this repository in your research, please consider citing the following references:

bibtex @article{neto2024approaches, title={Approaches based on language models for aspect extraction for sentiment analysis in the Portuguese language}, author={Neto, Jos{\'e} Carlos Ferreira and Pereira, Denilson Alves and Barbosa, Bruno Henrique Groenner and Ferreira, Danton Diego}, journal={Neural Computing and Applications}, pages={1--11}, year={2024}, publisher={Springer} }

📮 Contact

For questions or suggestions, please get in touch with us via email: eng.jcfneto@email.com.