RuBQ

RuBQ是由ITMO大学和JetBrains研究团队创建的首个俄语知识库问答数据集，包含1500个不同复杂度的俄语问题及其英文机器翻译、相应的SPARQL查询和答案。数据集的创建始于大量在线测验的问答对，经过自动过滤、众包实体链接、SPARQL查询自动生成和内部验证。该数据集适用于广泛的语义网、自然语言处理和信息检索领域的研究者和实践者，尤其是那些从事多语言问答系统的研究。

RuBQ is the first Russian knowledge base question answering (KBQA) dataset developed by research teams from ITMO University and JetBrains. It comprises 1500 Russian questions of varying complexity, paired with their machine translations into English, corresponding SPARQL queries, and reference answers. The dataset was constructed from question-answer pairs collected from a large number of online quizzes, followed by a processing pipeline including automatic filtering, crowdsourced entity linking, automatic SPARQL query generation, and internal validation. This dataset is suitable for researchers and practitioners across a wide range of fields including the Semantic Web, natural language processing (NLP), and information retrieval, especially those engaged in research on multilingual question answering systems.