多主题多方言阿拉伯语数据集

多主题多方言阿拉伯语数据集是由穆罕默德六世理工大学创建的开放访问数据集，包含来自Twitter的49,306条推文，涵盖五个阿拉伯国家方言。数据集内容丰富，涉及多个主题和情感分析，旨在解决阿拉伯语方言资源稀缺的问题。创建过程包括数据收集、清洗、标注等步骤，确保数据质量。该数据集适用于阿拉伯语方言的NLP任务，如方言识别、主题分类和情感分析，有助于推动相关领域的研究和应用。

The Multi-topic Multi-dialect Arabic Dataset is an open-access dataset developed by Mohammed VI Polytechnic University. It contains 49,306 tweets sourced from Twitter, covering dialects from five Arabic-speaking countries. The dataset features rich content spanning multiple topics and sentiment analysis tasks, aiming to address the scarcity of Arabic dialectal language resources. Its development workflow includes data collection, cleaning, annotation and other procedures to ensure data quality. This dataset is applicable to NLP tasks for Arabic dialects, such as dialect identification, topic classification and sentiment analysis, and can facilitate research and application advancement in relevant fields.