摘要
针对传统的K-means算法在过分依赖初始聚类中心选取方面的不足,论文提出了一种基于自适应PSO的K-means聚类算法。该算法设计了一种自适应惯性权重函数对PSO进行动态调整,然后与K-means算法融合,使K-means的各个初始聚类中心能自适应生成,达到全局最优,最后将上述改进的聚类算法应用于医学电子病历数据病症的聚类处理。实验结果表明该算法具有更高的电子病历病症聚类准确率和执行效率。
In view of the shortcomings of traditional K-means algorithm in over reliance on the initial cluster center selection,this paper presents a K-means clustering algorithm based on adaptive PSO.The algorithm designs an adaptive inertia weight function to dynamically adjust the PSO,and then converge with the K-means algorithm to make the initial clustering center of K-means adaptively generate the global optimal.Finally,the improved clustering algorithm is applied to the clustering of medical electronic medical records.The experimental results show that the algorithm has higher accuracy and efficiency when it is applied in cluster analysis of electronic medical records.
作者
沐燕舟
丁卫平
高峰
余利国
张琼
MU Yanzhou;DING Weiping;GAO Feng;YU Liguo;ZHANG Qiong(School of Computer Science and Technology,Nantong University,Nantong 226019)
出处
《计算机与数字工程》
2019年第8期1861-1865,共5页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:61300167)
江苏省自然科学基金项目(编号:BK20151274)
江苏省六大人才高峰项目(编号:XYDXXJS-048)
江苏省高校“青蓝工程”项目
南通市应用基础研究项目(编号:GY12016014)资助