Abstract:

Information explosion makes it difficult for patients to choose a doctor, and it has also limited the efficient use of Internet medical services. In this paper, a study is conducted on an intelligent medical recommendation model that considers the diversity of patient preferences for medical treatment, mainly focusing on the decision-making preferences of patients with sustained usage behavior, such as preferences for doctor service quality, knowledge contribution, electronic reputation, and medical experience, taking into account graded diagnostic and treatment standards when constructing feature variables. Based on the relative quantity, the comprehensive score is calculated by using the entropy weight-TOPSIS multi-index evaluation model, and the score result is used as the target variable, while the support vector regression algorithm optimized by the improved sparrow search algorithm is used to mine the relationship between the feature variable and the target variable, thereby constructing an intelligent medical recommendation model. According to the recommendation model, the comprehensive performance of doctors recommended is better, and the adoption rate of recommendations has also increased.

Key words:

medical selection, intelligent recommendation, sparrow search algorithm, support vector regression

摘要：

信息爆炸导致患者择医难，也限制了互联网医疗服务高效发挥作用。围绕考虑患者择医偏好多样性的智能就医推荐模型展开研究，重点关注具有持续使用行为患者的择医决策偏好特征，如对医生服务质量、知识贡献量、电子口碑、诊疗经验的偏好等，并将分级诊疗标准纳入考虑，构建特征变量。同时，以相对地位量数为基础，通过熵权-TOPSIS多指标评价模型计算综合评分，将评分结果作为目标变量，并利用改进麻雀搜索算法优化的支持向量回归算法挖掘特征变量与目标变量间的数据关系，从而构建智能就医推荐模型。根据该推荐模型所推荐的医生综合表现较好，推荐采纳率也有所提高。

关键词:

就医选择, 智能推荐, 麻雀搜索算法, 支持向量回归