基于舆情的智能问询函数据集

本数据集包含金融监管中关于舆情信息的问询函。为使用大模型的自动问询函生成提供了相应的数据支持，可用于大模型的指令微调阶段。数据集收集并整合了上海证券交易所中2016-2023年47个舆情相关的问询函，并人工检索获取了其对应的舆情文本，最终得到了43条样本构成的训练集和5条数据构成的测试集。This dataset comprises inquiry letters related to public opinion information in financial supervision. It provides targeted data support for the automatic generation of inquiry letters using large language models (LLMs), and is suitable for the instruction fine-tuning phase of LLMs. The dataset collects and consolidates 47 public opinion-related inquiry letters from the Shanghai Stock Exchange spanning 2016 to 2023, and manually retrieves their corresponding public opinion texts. Finally, a training set consisting of 43 samples and a test set consisting of 5 data entries are obtained.

搜集汇总

背景与挑战

背景概述

该数据集聚焦金融监管中的舆情问询函，旨在支持大模型的自动生成和指令微调。它整合了上海证券交易所2016-2023年的47个舆情相关问询函及其对应文本，最终构建了包含43条训练样本和5条测试样本的数据集。