高中语文主观题5类题型数据集

高中语文主观题5类题型数据集为高中区域联考、学校期中、期末考试中“文学类阅读”“文言文阅读”等数据集。每种题型2万条训练数据，按5个题型算，即10万条。数据包括：题目+学生回答+评分；每种题型覆盖100个以上的题目；学生回答的质量覆盖各分数段，高低分段占比不少于10%；学校质量覆盖当地教学水平的前中后。本类数据以条为单位。每条数据中涉及个人信息需以匿名化形式提供，题目和作答图片要求清晰可识别（清晰度大于300PPI，手写字迹可识别，内容识别率>99%）

The 5-type high school Chinese subjective question dataset is sourced from regional joint exams, mid-term and final term examinations of high schools, covering question types such as "literary text reading" and "classical Chinese text reading". Each question type contains 20,000 training samples, summing up to 100,000 samples in total across all 5 types. The dataset comprises three core components: questions, students' responses, and corresponding scores. Each question type includes more than 100 distinct questions. Students' responses cover all score ranges, with the proportions of high and low score segments each accounting for no less than 10%. The participating schools span the top, middle and bottom tiers of local teaching quality levels. This dataset takes individual samples as the basic unit. All personal information contained in each sample must be anonymized. The images of questions and students' answers are required to be clearly identifiable: the resolution must be higher than 300 PPI, handwritten text must be legible, and the content recognition rate should exceed 99%.