摘要
针对传统医疗数据趋势算法在面对海量数据时处理效率不高、准确率低、难以满足应用需求等问题,文中提出了一种基于改进朴素贝叶斯的大数据趋势预测算法。该算法首先对未加工的数据进行预处理来提高数据质量,然后在其基础上提取数据特征,提取完成后利用训练好的朴素贝叶斯分类算法模型对数据进行处理,并将处理结果作为K近邻算法的输入,经二次处理得到最终的趋势预测结果。实验结果表明,相较于其他同类趋势预测算法,所提算法能够在确保效率的前提下,使准确率稳定在92%以上,且能在多并发的情况下保持算法稳定。
In response to the problems of low processing efficiency,low accuracy,and difficulty in meeting application needs of traditional medical data trend algorithms in the face of massive data,this paper proposes a big data trend prediction algorithm based on improved naive Bayes.The algorithm firstly preprocesses the unprocessed data to improve data quality,and then extracts data features based on it.After the extraction is completed,the trained naive Bayesian algorithm model is used to process the data,and the processing results are used as input for the K-nearest neighbor algorithm for secondary processing to obtain the final trend prediction result.The experiment results show that compared to other similar trend prediction algorithms,the proposed algorithm can maintain an accuracy of over 92%while ensuring efficiency,and can maintain the stability of the algorithm in multi concurrency situations.
作者
岳晓磊
刘欣
YUE Xiao-lei;LIU Xin(The First Affiliated Hospital of Hebei North University,Zhangjiakou 075000,Hebei Province,China)
出处
《信息技术》
2025年第12期89-93,共5页
Information Technology
基金
河北省科技厅科技支撑计划项目(182777214)。
关键词
贝叶斯模型
朴素贝叶斯
K近邻算法
数据处理
趋势预测
Bayesian model
naive Bayes
K-nearest neighbor algorithm
data processing
trend prediction
