---
license: apache-2.0
dataset_info:
  features:
  - name: text
    dtype: string
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 71907240
    num_examples: 85027
  download_size: 43818509
  dataset_size: 71907240
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- pt
pretty_name: sentiment-analysis-pt
size_categories:
- 10K<n<100K
---

# Sentiment Analysis (Portuguese) dataset

This dataset is a concatenation of the [`IMDB 50K`](https://www.kaggle.com/datasets/lakshmi25npathi/imdb-dataset-of-50k-movie-reviews?select=IMDB+Dataset.csv), the [`Twitter US Airline Sentiment`](https://www.kaggle.com/datasets/crowdflower/twitter-airline-sentiment), App Reviews scraped from [Google Play](https://github.com/Nkluge-correa/teeny-tiny_castle/blob/master/ML%20Explainability/NLP%20Interpreter/text_scraping.ipynb), and the [`EcoPreprocessed`](https://www.kaggle.com/datasets/pradeeshprabhakar/preprocessed-dataset-sentiment-analysis).

- **Note:** Samples were translated via the Google Translate API.


## Overview 


## Dataset Details


**Citation:**
```latex

```
## Contents

The dataset consists of a data frame with the following columns:

-**text:**

-**label:**


## How to use

```python
from datasets import load_dataset

dataset = load_dataset("AiresPucrs/sentiment-analysis-pt", split='train')

```

## License

This dataset is licensed under the Apache License 2.0.