MWirelabs/mizo-language-corpus-4M

Mizo-Language-Corpus-4M是一个开源的单语种Mizo语言数据集，包含400万个句子，由MWireLabs精心制作，旨在支持自然语言处理研究，促进语言平等，并在低资源语言中推动开放开发。该数据集适用于训练和评估Mizo语言模型，包括掩码语言建模、情感分析、命名实体识别、文本分类、机器翻译等NLP任务。The Mizo-Language-Corpus-4M is an open-source monolingual Mizo dataset containing 4 million sentences, curated by MWireLabs to support natural language processing research, promote linguistic equity, and foster open development in low-resource languages. It is suitable for training and evaluating Mizo language models for tasks such as masked language modeling, sentiment analysis, named entity recognition, text classification, machine translation, and more.