README.md

---
task_categories:
- text-classification
language:
- vi
---
## VLSP2018-ABSA-Hotel

### Dataset Summary

The VLSP 2018 Hotel corpus is designed for Vietnamese Aspect-Based Sentiment Analysis (ABSA), covering two sub-tasks of Aspect Category Sentiment Analysis (ACSA):

1. **Aspect Category Detection (ACD):** identify which `Aspect#Category` pairs are present in each review.
2. **Sentiment Polarity Classification (SPC):** assign one of three sentiment labels (`Positive`, `Negative`, `Neutral`) to each detected `Aspect#Category`.

This unified CSV contains **5,600** reviews with a `type` column for `train`/`dev`/`test` and binary indicators for each aspect–category (0=no, 1=positive, 2=negative, 3=neutral).


### Supported Tasks and Metrics

* **Aspect Category Detection:** multi-label classification
* **Sentiment Polarity Classification:** multi-class classification
* **Metrics:** Precision, Recall, F1 (for both ACD and SPC)


### Languages

* Vietnamese


### Dataset Structure

| Column                           | Type  | Description                                               |
| -------------------------------- | ----- | --------------------------------------------------------- |
| `Review`                         | `str` | The raw hotel review text.                                |
| `<Aspect#Category>` (34 columns) | `int` | One-hot+polarity indicator per aspect#category (0/1/2/3). |
| `type`                           | `str` | Split: `train` / `validation` / `test`.                   |
| `dataset`                        | `str` | Always `VLSP2018-ABSA-Hotel`.                             |

The 34 aspect–category columns are:

```
FACILITIES#CLEANLINESS, FACILITIES#COMFORT, …, SERVICE#GENERAL
```

(each value: 0=absent, 1=positive, 2=negative, 3=neutral)


### Usage

```python
from datasets import load_dataset

ds = load_dataset("visolex/vlsp2018-absa-hotel")

train = ds.filter(lambda ex: ex["type"] == "train")
val   = ds.filter(lambda ex: ex["type"] == "dev")
test  = ds.filter(lambda ex: ex["type"] == "test")

print(train[0])
```


### Source & Links

* **GitHub:** [https://github.com/ds4v/absa-vlsp-2018](https://github.com/ds4v/absa-vlsp-2018 "https://github.com/ds4v/absa-vlsp-2018") 
* **Publication:** End-to-end Multi-task Solutions for ACSA on Vietnamese (VLSP 2018)


### Citation

```bibtex
@INPROCEEDINGS{9865479,
  author={Dang, Hoang-Quan and Nguyen, Duc-Duy-Anh and Do, Trong-Hop},
  booktitle={2022 IEEE International Conference on Cybernetics and Computational Intelligence (CyberneticsCom)}, 
  title={Multi-task Solution for Aspect Category Sentiment Analysis on Vietnamese Datasets}, 
  year={2022},
  volume={},
  number={},
  pages={404-409},
  keywords={Sentiment analysis;Analytical models;Computational modeling;Multitasking;Task analysis;Cybernetics;Computational intelligence;Aspect-based Sentiment Analysis;PhoBERT;Aspect Category Detection;Sentiment Polarity Classification},
  doi={10.1109/CyberneticsCom55287.2022.9865479}}
}
```