Bangla Book Recommendation Dataset

Summary

This repository contains the dataset and code for the paper Towards Personalized Bangla Book Recommendation: A Large-Scale Multi-Entity Book Graph Dataset. In this work, we introduce a large-scale multi-entity heterogeneous graph dataset for Bangla book recommendation, integrating users, books, authors, publishers, and categories. Our framework enables the development of sophisticated graph-based recommendation systems. This contribution aims to enhance the discovery of Bangla literature through more accurate and personalized user experiences.

---

Models (Google Colab)

Click any badge below to open and run instantly in Google Colab.
(Dataset downloads automatically — no setup needed!)

Model	Open in Colab	Model	Open in Colab
Global Popularity		Category-Aware Popularity
User-Based Collaborative Filtering		Item-Based Collaborative Filtering
Implicit Matrix Factorization (ALS)		Explicit Matrix Factorization (SVD)
Two Tower Text Embedding		LightGCN
Pure Content Based

---

Run Locally (Anaconda / Jupyter Notebook)

Prefer running on your own machine? The setup guide walks you through everything — conda environment, library installation, and model-specific dependencies (including the tricky DGL install for HGNN).

---

Citation

If you use the code, dataset or model checkpoints, please cite the following work:

@misc{ahmed2026personalizedbanglabookrecommendation,
      title={Towards Personalized Bangla Book Recommendation: A Large-Scale Multi-Entity Book Graph Dataset}, 
      author={Rahin Arefin Ahmed and Md. Anik Chowdhury and Sakil Ahmed Sheikh Reza and Devnil Bhattacharjee and Muhammad Abdullah Adnan and Nafis Sadeq},
      year={2026},
      eprint={2602.12129},
      archivePrefix={arXiv},
      primaryClass={cs.IR},
      url={https://arxiv.org/abs/2602.12129}, 
}