Application of multilayer perceptron neural networks and support vector machines in classification of healthcare data
Application of multilayer perceptron neural networks and support vector machines in classification of healthcare data
复制标题
多层感知器神经网络和支持向量机在医疗数据分类中的应用
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Alireza Sadeghian
中科院分区：
文献类型：
--
作者：
Parisa Naraei;A. Abhari;Alireza Sadeghian
医疗保健行业每天都在稳定地产生大量数据。数据挖掘和机器学习方法是两种有效的技术，适用于数据分析和发现隐藏的模式，可用于医疗决策。由于医疗领域的决策是处理病人的结果，在数据挖掘的准确性高的水平是必要的。本文在心脏病数据集上对多层感知器神经网络和支持向量机进行了比较。我们分析了支持向量机的有效性分类，使用303例患者的数据集。我们的结果表明，支持向量机是能够更准确地分类。
A large volume of data is steadily produced by the healthcare industry on daily basis. Data mining and machine learning approaches are two effective techniques applicable for data analysis and finding the hidden patterns which can be utilized for medical decision making. As the decisions in medical field are dealing with patient outcome, a high level of accuracy in data mining is needed. In this paper a comparison between implemented multilayer perceptron neural networks and support vector machine on heart diseases dataset is conducted. We have analyzed the effectiveness of support vector machine in classification, using a dataset of 303 patients. Our results show that support vector machine is able to classify more accurately.