Books Dataset

This dataset supports the research article titled <b>“A Multi-Label Classification Framework for Semantic and Contextual Label Prediction from Textual Data.”</b> The dataset is used to evaluate traditional similarity-based methods, machine learning models, and large language model (LLM)-based pipelines for multi-label text classification.The data is derived from an openly available books dataset and includes structured and processed textual information required to reproduce the experimental results reported in the manuscript. It contains book titles, descriptions, author information, publication metadata, and multi-label category assignments used for semantic and contextual classification tasks.This dataset was utilized to generate all quantitative results, tables, and figures presented in the paper, including precision, recall, and F1-score evaluations across multiple Top-N prediction scenarios. Preprocessed versions of the data, label mappings, and derived files used during experimentation are included to ensure full reproducibility of the study.The dataset is intended for researchers and practitioners working in multi-label text classification, recommender systems, semantic analysis, and natural language processing. All data is shared for research and academic purposes under an open-access license, enabling reuse, validation, and extension of the proposed SmarTClass framework.