metadata

viewer: true
task_categories:
  - text-classification
language:
  - vi
tags:
  - sentiment
size_categories:
  - 10K<n<100K

Vietnamese Comment Sentiment Dataset

Overview

This dataset contains Vietnamese comments collected from various social networks to facilitate sentiment analysis. Each comment is labeled to indicate its sentiment, making it useful for natural language processing tasks.

Data Source

The comments were crawled from several social network platforms, ensuring a diverse range of expressions and contexts within Vietnamese language usage.

Structure

The dataset is in CSV format, comprising the following columns:

ID: A unique identifier for each comment.
Title: The title associated with the comment.
Content: The full text of the comment.
BriefContent: A brief excerpt of the comment.
URL: The URL of the source from which the comment was collected.
Published Date: The date when the comment was published.
Week: The week of the year the comment corresponds to.
Keyword: Keywords associated with the content.
Group: Group classification of the comment.
Sub: Sub-category relating to the group.
Sentiment: The sentiment label, categorized as positive, negative, or neutral.

Usage

You can load and use the Vietnamese Comment Sentiment dataset from the Hugging Face Hub using the datasets library in Python. Here’s how you can do it:

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("minhtoan/vietnamese-comment-sentiment")

# Access the train split
train_data = dataset['train']

# Display a sample data point
print(train_data[0])

Author

dataset{minhtoan/vietnamese-comment-sentiment, title={Vietnamese Comment Sentiment Dataset}, author={Phan Minh Toan}, year={2022}, publisher={Hugging Face} }