Papers › BanglaBook: A Large-scale Bangla Dataset for Sentiment Analysis from Book Reviews
BanglaBook: A Large-scale Bangla Dataset for Sentiment Analysis from Book Reviews
Mohsinul Kabir, Obayed Bin Mahfuz, Syed Rifat Raiyan, Hasan Mahmud, Md Kamrul Hasan
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.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Sentiment Analysis | BanglaBook | Bangla-BERT (large) | Weighted Average F1-score | 0.9331 | #1 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Random Forest (word 2-gram + word 3-gram) | Weighted Average F1-score | 0.9106 | #2 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Bangla-BERT (base-uncased) | Weighted Average F1-score | 0.9064 | #3 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | SVM (word 2-gram + word 3-gram) | Weighted Average F1-score | 0.9053 | #4 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Random Forest (word 1-gram) | Weighted Average F1-score | 0.9043 | #5 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Logistic Regression (char 2-gram + char 3-gram) | Weighted Average F1-score | 0.8978 | #6 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Logistic Regression (word 2-gram + word 3-gram) | Weighted Average F1-score | 0.8964 | #7 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | XGBoost (char 2-gram + char 3-gram) | Weighted Average F1-score | 0.8723 | #8 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Multinomial NB (word 2-gram + word 3-gram) | Weighted Average F1-score | 0.8663 | #9 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | XGBoost (word 2-gram + word 3-gram) | Weighted Average F1-score | 0.8651 | #10 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | Multinomial NB (BoW) | Weighted Average F1-score | 0.8564 | #11 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | SVM (word 1-gram) | Weighted Average F1-score | 0.8519 | #12 of 13 | Archive leaderboard | report |
| Sentiment Analysis | BanglaBook | LSTM (GloVe) | Weighted Average F1-score | 0.0991 | #13 of 13 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections