Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Languages:
Persian
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
@@ -16,4 +16,68 @@ configs:
|
|
16 |
data_files:
|
17 |
- split: train
|
18 |
path: data/train-*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
data_files:
|
17 |
- split: train
|
18 |
path: data/train-*
|
19 |
+
license: apache-2.0
|
20 |
+
task_categories:
|
21 |
+
- text-classification
|
22 |
+
language:
|
23 |
+
- fa
|
24 |
+
size_categories:
|
25 |
+
- 1K<n<10K
|
26 |
---
|
27 |
+
|
28 |
+
# Dataset Card for Persian Sentiment Comments
|
29 |
+
|
30 |
+
## Dataset Description
|
31 |
+
|
32 |
+
**`persian-sentiment-comments`** is a curated dataset of Persian (Farsi) user comments, each annotated with a sentiment label. The dataset is designed for **sentiment analysis tasks** and is particularly suitable for training and evaluating machine learning models on Persian text data.
|
33 |
+
|
34 |
+
The comments are collected from a variety of sources, including online product reviews and user feedback. Each comment is paired with a sentiment label:
|
35 |
+
|
36 |
+
- **1:** Positive sentiment
|
37 |
+
- **0:** Negative sentiment
|
38 |
+
- *(Optionally, if your dataset includes neutral comments, specify as well)*
|
39 |
+
|
40 |
+
The dataset aims to support research and development in **Natural Language Processing (NLP)** for low-resource languages, especially for Persian sentiment classification.
|
41 |
+
|
42 |
+
### Example
|
43 |
+
|
44 |
+
| Text | Label |
|
45 |
+
|------|-------|
|
46 |
+
| این اولین تجربه من برای خرید ایفون هست امروز ب دستم رسیده به اسم طلایی هست ولی دراصل رزگلده خیلی هم خوش رنگ هست | 1 |
|
47 |
+
| خرید این محصول رو توصیه میکنم | 1 |
|
48 |
+
| یک ساله این گوشی رو دارم هیچ نقطه ضعفی ازش ندیدم | 1 |
|
49 |
+
| سلام خدمت دوستان این گوشی از همه نظر عالی کیفیت ساخت سرعت اجرا کیفیت صدا و تصویر از نظر حافظه هم فقط 5 گیگ در دسترس شما قرار میگیره | 1 |
|
50 |
+
| سلام دوستانی که نگران شکستن صفحه نمایش هستند این رو بدونند که صفحه با گوریلا گلس 5 محافظت میشه و با خدمات ویژه سامسونگ اگر تحت هر شرایط صفحه گوشی دچار ایراد بشه با 200 هزار تومان صفحه رو براتون تعویض میکنند و از لحاظ امکانات هم گوشی عالیه و من از خرید خودم راضیم | 1 |
|
51 |
+
|
52 |
+
## Dataset Structure
|
53 |
+
|
54 |
+
- **Text** (`string`): The Persian comment or review.
|
55 |
+
- **Label** (`int64`): The sentiment label (1 = positive, 0 = negative).
|
56 |
+
|
57 |
+
## Supported Tasks and Leaderboards
|
58 |
+
|
59 |
+
- **Sentiment Analysis**: Classification of Persian text into positive or negative sentiment.
|
60 |
+
- Suitable for both academic research and real-world applications in Persian NLP.
|
61 |
+
|
62 |
+
## Languages
|
63 |
+
|
64 |
+
- **Primary:** Persian (Farsi)
|
65 |
+
|
66 |
+
## Citation
|
67 |
+
|
68 |
+
If you use this dataset, please cite as:
|
69 |
+
|
70 |
+
```
|
71 |
+
@dataset{persian_sentiment_comments_2025,
|
72 |
+
title = {Persian Sentiment Comments Dataset},
|
73 |
+
author = {AICT Sharif},
|
74 |
+
year = {2025},
|
75 |
+
description = {A dataset of Persian user comments labeled for sentiment analysis.},
|
76 |
+
url = {https://huggingface.co/datasets/your-username/persian-sentiment-comments}
|
77 |
+
}
|
78 |
+
```
|
79 |
+
|
80 |
+
## Contributors
|
81 |
+
|
82 |
+
- [Soheil Tehranipour](https://huggingface.co/soheiltp)
|
83 |
+
- [Masoud Kaviani](https://huggingface.co/masoudkaviani)
|