摘要
研究高血压诊断预测准确性问题,需要根据病人的生活习性、体质指数、腰臀围以及病理特征,以完成对高血压的病理预测。但在预测过程中,由于个人的体质指数和病理特征很难保证充足,往往会出现数据属性单值突变的现象,造成对高血压预测准确率不高。提出采用大数据分析的潜在高血压病预测方法。通过采集相关数据信息,然后对采集到的数据进行包括数据清理、数据转化和数据集成的预处理,根据支持向量机理论,对数据属性进行分类,建立潜在高血压病的预测模型,计算模型属性分类结果的权重,得到不同属性对高血压病影响的重要程度分级,通过与高血压病特征参数的比较,获取潜在高血压病的预测结果。实验结果表明,采用改进算法进行潜在高血压病的预测,能够有效提高预测的准确率与预测效率,为早期高血压病的检测与防治提供数据保障,进而满足医学检测的实际需求。
A prediction method for potential hypertension based on big data analysis is proposed. By collecting relevant data information and carrying out pretreatment for collected data, including data cleaning, data transformation and data integration, according to the theory of support vector machine (SVM) to classify data attributes, a prediction model for potential hypertension is established, the weights of model attribute classification results are calculated, and an importance classification of different attribute for hypertension is obtained. Compared with hypertension characteristic parameters, the potential prediction result of hypertension is obtained. The experimental results show that the proposed algorithm can effectively improve the prediction accuracy and efficiency, and provide data security for the early detection and prevention of potential hypertension.
出处
《计算机仿真》
CSCD
北大核心
2015年第5期386-389,421,共5页
Computer Simulation
基金
辽宁省教育厅科研课题(L2013357)
关键词
大数据分析
潜在高血压病预测
支持向量机
Large data analysis
Potential hypertension prediction
Support vector machine (SVM)