Diabetes Disease Prediction Using Machine Learning on Big Data of Healthcare
Diabetes Disease Prediction Using Machine Learning on Big Data of Healthcare
复制标题
利用医疗保健大数据的机器学习来预测糖尿病
DOI：
10.1109/iccubea.2018.8697439
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Sudhir N. Dhage
中科院分区：
文献类型：
--
作者：
A. Mir;Sudhir N. Dhage
医疗保健领域是一个技术进步日新月异、数据量日益增长的非常突出的研究领域。为了处理海量的医疗数据，我们需要大数据分析，这是医疗领域的一种新兴方法。数以百万计的患者在全球范围内通过各种程序寻求治疗。分析用于诊断特定疾病的患者的治疗趋势将有助于做出知情和有效的决策，以提高医疗保健的整体质量。机器学习是一种非常有前景的方法，它有助于疾病的早期诊断，并可能帮助实践者做出诊断决策。本文旨在利用朴素贝叶斯、支持向量机、随机森林和简单CART算法，利用WEKA工具建立预测糖尿病疾病的分类器模型。本研究希望基于高效的性能结果推荐最优的糖尿病疾病预测算法。对每种算法在数据集上的实验结果进行了评估。可以观察到，支持向量机在疾病预测中表现得最好，具有最高的精度。
Healthcare domain is a very prominent research field with rapid technological advancement and increasing data day by day. In order to deal with large volume of healthcare data we need Big Data Analytics which is an emerging approach in Healthcare domain. Millions of patients seek treatments around the globe with various procedure. Analyzing the trends in treatment of patients for diagnosis of a particular disease will help in making informed and efficient decisions to improve the overall quality of healthcare. Machine Learning is a very promising approach which helps in early diagnosis of disease and might help the practitioners in decision making for diagnosis. This paper aims at building a classifier model using WEKA tool to predict diabetes disease by employing Naive Bayes, Support Vector Machine, Random Forest and Simple CART algorithm. The research hopes to recommend the best algorithm based on efficient performance result for the prediction of diabetes disease. Experimental results of each algorithm used on the dataset was evaluated. It is observed that Support Vector Machine performed best in prediction of the disease having maximum accuracy.