摘要
支持向量机(SVM)普遍应用在机器学习领域的学习算法,广泛用于分类学习。支持向量机也应用在很多实际应用领域中。该算法也广泛地应用在煤炭系统的分类预测工作中。随着数字时代的发展,煤炭系统的数据规模也呈现大规模增长趋势。针对海量规模数据,传统的支持向量机模型不能有效地完成煤炭系统中数据的分类、回归等工作。文章针对大规模数据处理困难的问题,提出了分布式支持向量机模型。该模型针对现有流行的云计算平台,在该平台下构建基于Hadoop分布式计算框架的分布式模型,该分布式支持向量机模型能够高效、快速地完成真实数据的分类或回归任务,具有很高的效率。文中的实验部分通过大量的实验数据进一步证明了文章提出算法的可行性。
Support vector machine (SVM) is a learning algorithm which is widely applied in machine learning area, and used for classification learning. And, SVM is also applied into other areas. This algorithm can be used into forecast work in coal system. With the development of digital age, the data in coal system is increasing with big scale trend. Focusing on huge scale data, traditional support vector machine model could not complete the data classification or regression work in coal system. In this paper, focusing on the difficult problem of dealing with large scale data, we propose distributed support vector machine model. This model is based on popular cloud computing platform, which is based on Hadoop distributed computing framework. This distributed support vector machine model could complete the classification or regression task for the real data effectively and fast, and it has high efficiency. The experiments in this paper prove the feasibility of our proposed algorithm through numerous of experimental data.
出处
《煤炭技术》
CAS
北大核心
2013年第11期248-250,共3页
Coal Technology
关键词
支持向量机
云计算
分布式
support vector machine cloud computing hadoop classification distributed