【摘要】 目的 基于数据挖掘探讨耳穴贴压治疗痤疮的选穴规律。方法 检索中国知网、万方数据库、维普数据库、中国生物医学文献服务系统和PubMed建库以来至2021年11月8日关于耳穴贴压治疗痤疮患者的相关文献，运用NoteExpress进行筛选，Excel 2019建立数据库信息，SPSS 23.0及SPSS Modeler 18.0进行因子分析、聚类分析及关联规则。结果 68篇文献分析耳穴使用频次最高的依次为内分泌、肺、面颊，聚类分析显示2大类（皮质下-神门-内分泌-面颊-交感；脾-肝-大肠-肺-胃-肾上腺）和5小类（皮质下-神门-内分泌；面颊-交感；脾-肝-大肠；肺-胃；肾上腺），因子分析提取出5个公因子（脾、肝、大肠；皮质下、神门；面颊、肾上腺；内分泌、交感；肺），关联规则得出耳穴贴压皮质下与神门、面颊与胃、肺与胃、肺与脾等穴位配伍意义较大。结论 通过数据挖掘得出具有指导性意义的4组穴位配伍，5个有效聚类群，5个公因子，可为耳穴选穴治疗痤疮提供参考依据。更多还原

【Abstract】 Objective To explore the rule of auricular point selection for sticking therapy of acne based on data mining. Methods Relevant literaturs pressure in applied to ear for managing acne was searched and screened from such databases of CNKI, Wanfang database, VIP database, Chinese biomedical literature service system and PubMed from the inception to 8th November 2021 by using NoteExpress. Excel 2019 was used to establish a database, which was analyzed using SPSS 23.0 and SPSS Modeler 18.0 for factor analysis, cluster analysis and association rules. Results A total of 68 papers that met the search criteria were included for the analysis. The auricular acupoints selected for acne management included those corresponding to endocrine, lung and cheek. Cluster analysis showed that there were two categories(subcortical-Shenmen-endocrine-cheek-sympathia; spleen-liver-large intestine-lung-stomach-adrenal gland) and five subcategories, e.g., subcortical-divine-endocrine, cheek-sympathia, spleen-liver-large intestine, lung-stomach and adrenal gland. By factor analysis the five common factors were extracted including spleen-liver-large-intestine, subcortical-gate, cheeks-adrenal gland, endocrine-sympathia and lung. According to the rules of association, it was concluded that ear acupuncture pressure under the cortex played a great significant role in acupoint compatibility with the acupoints of Shenmen, cheek and stomach, lung and stomach, lung and spleen.Conclusion Through data mining, the inventories significant for acne management are four groups of acupoint compatibility, five effective clustering groups and five common factors.更多还原