Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,77 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-nc-sa-4.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-sa-4.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- text-classification
|
| 5 |
+
- zero-shot-classification
|
| 6 |
+
language:
|
| 7 |
+
- bn
|
| 8 |
+
tags:
|
| 9 |
+
- Sentiment Analysis
|
| 10 |
+
- Book Reviews
|
| 11 |
+
- Product Reviews
|
| 12 |
+
- Bangla
|
| 13 |
+
- Bengali
|
| 14 |
+
- Dataset
|
| 15 |
+
pretty_name: BanglaBook
|
| 16 |
+
size_categories:
|
| 17 |
+
- 100K<n<1M
|
| 18 |
+
---
|
| 19 |
+
# BᴀɴɢʟᴀBᴏᴏᴋ: A Large-scale Bangla Dataset for Sentiment Analysis from Book Reviews
|
| 20 |
+
This repository contains the code, data, and models of the paper titled "BᴀɴɢʟᴀBᴏᴏᴋ: A Large-scale Bangla Dataset for Sentiment Analysis from Book Reviews" published in the ***Findings of the Association for Computational Linguistics: ACL 2023***.
|
| 21 |
+
|
| 22 |
+
[](https://arxiv.org/abs/2305.06595)
|
| 23 |
+
[](https://aclanthology.org/2023.findings-acl.80/)
|
| 24 |
+
[](https://tinyurl.com/gscholarbanglabook)
|
| 25 |
+
[](https://www.researchgate.net/publication/370688086_BanglaBook_A_Large-scale_Bangla_Dataset_for_Sentiment_Analysis_from_Book_Reviews)
|
| 26 |
+
|
| 27 |
+
[](https://aclanthology.org/2023.findings-acl.80.pdf)
|
| 28 |
+
[](https://drive.google.com/file/d/1-UkYs_Rx11S7qKOfR-6rnO2VDp3W78vQ/view?usp=sharing)
|
| 29 |
+
[](https://aclanthology.org/2023.findings-acl.80.mp4)
|
| 30 |
+
|
| 31 |
+
**License:** Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
|
| 32 |
+
|
| 33 |
+
[](http://creativecommons.org/licenses/by-nc-sa/4.0/)
|
| 34 |
+
|
| 35 |
+
## Data Format
|
| 36 |
+
Each row consists of a book review sample. The table below describes what each column signifies.
|
| 37 |
+
|
| 38 |
+
Column Title | Description
|
| 39 |
+
------------ | -------------
|
| 40 |
+
`id` | The unique identification number of the sample
|
| 41 |
+
`Book_Name` | The title of the book that has been evaluated by the review
|
| 42 |
+
`Writer_Name` | The name of the book's author
|
| 43 |
+
`Category` | The genre to which the book belongs
|
| 44 |
+
`Rating` | A numerical value \\(r\\) such that \\(1\leq r \leq 5\\)<br>A score reflecting the reviewer's subjective assessment of the book's quality
|
| 45 |
+
`Review` | The review text written by the reviewer
|
| 46 |
+
`Site` | The name of the online bookshop
|
| 47 |
+
`sentiment` | The conveyed sentiment and class label of the review<br>For a review sample \\(i\\) with rating \\(r_i\\), the sentiment label \\(S_i\\) is,<br>
|
| 48 |
+
$$
|
| 49 |
+
S_i =\begin{cases}
|
| 50 |
+
\text{Negative}, & \text{if } r_i \leq 2\\
|
| 51 |
+
\text{Neutral}, & \text{if } r_i = 3\\
|
| 52 |
+
\text{Positive}, & \text{if }r_i \geq 4
|
| 53 |
+
\end{cases}
|
| 54 |
+
$$
|
| 55 |
+
`label` | The numerical representation of the sentiment label<br>For a review sample \\(i\\) with sentiment label \\(S_i\\), the numerical label is,<br>
|
| 56 |
+
$$label_i = \begin{cases} 0, &\text{if } S_i = \text{Negative} \\ 1, &\text{if } S_i = \text{Neutral} \\ 2, &\text{if } S_i = \text{Positive} \\ \end{cases}$$
|
| 57 |
+
|
| 58 |
+
## Citation
|
| 59 |
+
If you find this work useful, please cite our paper:
|
| 60 |
+
```bib
|
| 61 |
+
@inproceedings{kabir-etal-2023-banglabook,
|
| 62 |
+
title = "{B}angla{B}ook: A Large-scale {B}angla Dataset for Sentiment Analysis from Book Reviews",
|
| 63 |
+
author = "Kabir, Mohsinul and
|
| 64 |
+
Bin Mahfuz, Obayed and
|
| 65 |
+
Raiyan, Syed Rifat and
|
| 66 |
+
Mahmud, Hasan and
|
| 67 |
+
Hasan, Md Kamrul",
|
| 68 |
+
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
|
| 69 |
+
month = jul,
|
| 70 |
+
year = "2023",
|
| 71 |
+
address = "Toronto, Canada",
|
| 72 |
+
publisher = "Association for Computational Linguistics",
|
| 73 |
+
url = "https://aclanthology.org/2023.findings-acl.80",
|
| 74 |
+
pages = "1237--1247",
|
| 75 |
+
abstract = "The analysis of consumer sentiment, as expressed through reviews, can provide a wealth of insight regarding the quality of a product. While the study of sentiment analysis has been widely explored in many popular languages, relatively less attention has been given to the Bangla language, mostly due to a lack of relevant data and cross-domain adaptability. To address this limitation, we present BanglaBook, a large-scale dataset of Bangla book reviews consisting of 158,065 samples classified into three broad categories: positive, negative, and neutral. We provide a detailed statistical analysis of the dataset and employ a range of machine learning models to establish baselines including SVM, LSTM, and Bangla-BERT. Our findings demonstrate a substantial performance advantage of pre-trained models over models that rely on manually crafted features, emphasizing the necessity for additional training resources in this domain. Additionally, we conduct an in-depth error analysis by examining sentiment unigrams, which may provide insight into common classification errors in under-resourced languages like Bangla. Our codes and data are publicly available at https://github.com/mohsinulkabir14/BanglaBook.",
|
| 76 |
+
}
|
| 77 |
+
```
|