sixfingerdev/turkish-qa-multi-dialog-dataset

该数据集包含两个主要部分：约19,000个问答（QA）示例和由多步、自然土耳其语对话组成的对话数据。QA部分包含类似SQuAD结构的输入-输出示例，每行包含一个问题和明确的答案。对话部分包含真实的土耳其语对话示例，保留了用户和助手的角色，并包括访谈、日常聊天、技术咨询等多种自然场景。该数据集适用于通用土耳其语QA模型和聊天/聊天机器人模型。数据集以JSON行格式存储，可用于训练土耳其语QA模型、微调土耳其语LLM等多种用途。

This dataset contains two main parts: approximately 19,000 question-answer (QA) examples and multi-step, natural Turkish dialogue data. The QA section includes SQuAD-like transformed input-output examples, with each line featuring a single question and a clear answer. The dialogue section contains real Turkish conversation examples, preserving user and assistant roles, and includes various natural scenarios like interviews, casual chats, and technical consultations. The dataset is suitable for general-purpose Turkish QA models and chat/chatbot models. The data is stored in JSON lines format and can be used for various purposes such as training Turkish QA models, fine-tuning Turkish LLMs, and more.