地球资源数据云——数据资源详情

医疗保健症状 - 疾病分类数据集

该数据集《Healthcare Symptoms–Disease Classification Dataset》主要用于多分类任务，数据形态以文本为主，应用场景偏向医疗健康。 题目说明：A synthetic medical dataset mapping patient symptoms to 30 diseases for machine 任务类型：文本多分类。 建议流程：先做文本清洗与分词，再比较 TF - IDF+线性模型 与 预训练语言模型。 评估建议：使用分层切分或交叉验证，优先关注 F1、Recall、AUC 等分类指标。 可用文件：Healthcare.csv。 This dataset contains 25,000 synthetic healthcare records designed for machine learning models that classify diseases based on patient symptoms. It includes demographic attributes, symptom lists, and confirmed diagnoses across 30 common acute, chronic, infectious, and neurological diseases. The dataset is well - suited for: Multi - class disease classification Symptom pattern analysis Medical decision support modeling NLP feature extraction on symptom text Data mining and biomedical research Each record corresponds to a unique patient with a generated combination of symptoms and diagnosis created from realistic patterns while maintaining anonymity. This dataset is purely synthetic, meaning no real patient data is used.