Aesthetics Text Corpus and Comprehensive Stop Lemma List

本研究创建了名为‘Aesthetics Text Corpus’的印度语数据集，包含约1000篇跨越百年的虚构与非虚构文本。此外，还构建了一个包含超过200,000个词根的全面停止词根列表。数据集的创建旨在解决自然语言处理中非英语语言资源的不足，并强调使用停止词根而非停止词的重要性，因为词根在不同来源中更为一致。该数据集适用于语言模型创建、文本分类和信息检索等NLP任务，旨在推动印度语在NLP领域的应用。

This study developed an Indian-language dataset named 'Aesthetics Text Corpus', which contains approximately 1,000 fictional and non-fictional texts spanning over a century. Additionally, a comprehensive stop root list comprising more than 200,000 roots was constructed. The dataset was created to address the scarcity of non-English language resources in natural language processing (NLP), and to emphasize the critical importance of utilizing stop roots instead of stop words, as roots exhibit greater consistency across diverse source materials. This dataset is applicable to a variety of NLP tasks such as language model development, text classification, and information retrieval, with the goal of advancing the application of the Indian language in the NLP domain.