Update README.md
Browse files
README.md
CHANGED
@@ -20,7 +20,31 @@ dataset_info:
|
|
20 |
num_examples: 851
|
21 |
download_size: 50837
|
22 |
dataset_size: 109030
|
|
|
|
|
|
|
|
|
23 |
---
|
24 |
-
#
|
25 |
|
26 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
num_examples: 851
|
21 |
download_size: 50837
|
22 |
dataset_size: 109030
|
23 |
+
task_categories:
|
24 |
+
- text-classification
|
25 |
+
size_categories:
|
26 |
+
- n<1K
|
27 |
---
|
28 |
+
# darija-reviews
|
29 |
|
30 |
+
|
31 |
+
**Description**: This dataset small consists of product and service reviews from social media written in Darija, encompassing both Arabic and Arabizi writing styles.
|
32 |
+
Each review is categorized by its polarity (positive, negative, or neutral) and includes information about the domain or topic, as well as the writing style of the content.
|
33 |
+
The dataset encompasses a broad spectrum of topics, including clothing, cosmetics, entertainment, hospitality, IT, and other domains.
|
34 |
+
|
35 |
+
**Size**: The dataset contains a total of 851 reviews.
|
36 |
+
|
37 |
+
**Labels**:
|
38 |
+
|
39 |
+
- Sentiment Labels: Positive, Negative, Neutral
|
40 |
+
- Topics: Automotive, Cleaning, Clothing, Cosmetics, Entertainment, Hospitality, Household Appliances, IT, Jewelry, Restaurants, and Other.
|
41 |
+
- Writing Styles: Arabic & Arabizi
|
42 |
+
|
43 |
+
**Usage**: This dataset can be used for ***evaluating*** the performance of sentiment analysis models in classifying the polarity (positive, negative, or neutral) of product and service reviews written in Darija.
|
44 |
+
|
45 |
+
|
46 |
+
**Examples**:
|
47 |
+
|
48 |
+
- Positive IT Review (Arabic WS): "هاتف زوين عجبني"
|
49 |
+
- Negative Cosmetics Review (Arabizi WS): "Lwa7iid li fach kandirou wejhi kayt9lab blhboub manhar 9te3to ou lhbob hbso hada maysla7ch lbchra dohniya myidem bzzaf"
|
50 |
+
- Neutral Cleaning Review (Arabic WS): "خليه شوية على الطبيع ومسحي بشيفون فازك"
|