摘要
随着医疗卫生信息化建设的不断深入,医疗数据无论在种类上,亦或者是在规模上,都呈上升态势。而这些海量数据往往是高维的,因此,此次研究针对高维数据降维问题,设计了一种改进的二进制蚁群优化算法模型。首先,对基于二进制蚁群优化算法的特征选择算法进行研究,然后,利用该算法的特征权重进行改进,以降低高维数据的维度。对改进后的算法进行性能测试,当最大迭代次数达到50时,该算法性能就趋于平稳,性能优于对比算法。之后对研究提出的高维数据降维模型进行时间成本开销对比实验,结果显示,研究提出的高维数据降维模型相比对比模型降低的时间成本约为62%。上述实验结果表明,研究提出了一种改进的二进制蚁群优化算法模型,该模型在减少数据维度的同时,能够保持数据结构的有效性,提高数据挖掘的性能和效率。
With the deepening of medical and health information construction,medical data is on the rise both in type and scale.These massive data are often high-dimensional,so this study designed an improved binary ant colony optimization algorithm model for the problem of dimensionality reduction of high-dimensional data.Firstly,examination of the BACO-based feature selection process is being conducted,and then the feature weight of BACO algorithm is improved to reduce the dimension of high dimensional data.The performance of the improved BACO algorithm is tested.When the maximum number of iterations reaches about 50,the performance of the algorithm tends to be stable,and the performance is better than that of similar algorithms BACO and LCBBACO.A time cost comparison experiment was conducted on the high dimensional data reduction model based on RBFACO algorithm and BACO algorithm.Compared with the BACO high dimensional data reduction model on the Chronic Kidney Disease dataset,RBFACO reduced the time cost by about 62%.The above experimental results show that an improved binary ant colony optimization algorithm model is proposed,which can maintain the validity of data structure and improve the performance and efficiency of data mining while reducing the data dimension.
作者
罗康
刘帮富
李嘉
LUO Kang;LIU Bangfu;LI Jia(Baise City Maternal and Child Health Care Hospital,Baise,Guangxi 533000,China)
出处
《自动化与仪器仪表》
2024年第4期82-86,共5页
Automation & Instrumentation
基金
广州市天河区科技计划项目(201701YG014)。
