摘要
针对医院医疗投诉数据的复杂性,研究基于机器学习方法,旨在提升医院投诉处理效率和患者满意度。采用DBSCAN聚类算法识别相似投诉者群体,运用季节性分解时间序列分析投诉数量的波动,并利用分类模型进行效果预测。结果表明,聚类分析揭示了不同投诉者特征;季节性分析显示了投诉数量存在周期性波动。在预测投诉解决效果方面,优化后的XGBoost模型准确率达89%,F1分数为86%,显著优于其他模型。该研究为医院优化投诉处理流程提供了实证依据,有助于提升医疗服务质量。
In response to the complexity of hospital medical complaint data,this study aims to improve the efficiency of hospital complaint handling and patient satisfaction through machine learning methods.Using the DBSCAN clustering algorithm to identify similar groups of complainants,analyzing the fluctuations in the number of complaints using seasonal decomposition time series,and predicting the effectiveness using a classification model.The results indicate that cluster analysis reveals the characteristics of different complainants;Seasonal analysis shows periodic fluctuations in the number of complaints.In terms of predicting the effectiveness of complaint resolution,the optimized XGBoost model has an accuracy rate of 89% and an F1 score of 86%,significantly better than other models.The study provides empirical evidence for optimizing the complaint handling process in hospitals,which helps to improve the quality of medical services.
作者
陈威震
刘识
张智虎
刘丽红
韩霜雪
CHEN Weizhen;LIU Shi;ZHANG Zhihu;LIU Lihong;HAN Shuangxue(Doctor-patient Relationship Coordination Office of Beijing Tongren Hospital Affiliated to Capital Medical University,Beijing 100730,China)
出处
《电子设计工程》
2025年第20期136-140,共5页
Electronic Design Engineering
关键词
医疗投诉
机器学习
多维度评价
医疗服务质量
medical complaints
machine learning
multi dimensional evaluation
quality of medical services
