...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of Residuals Science & Technology >Application of Improved Decision Tree C4.5 Classification Algorithm in High - speed Data Stream Integrated Classification
【24h】

Application of Improved Decision Tree C4.5 Classification Algorithm in High - speed Data Stream Integrated Classification

机译：改进决策树C4.5分类算法在高速数据流综合分类中的应用。

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Data stream mining requires fast processing of data and adaptive concept drift in the context of occupying a small amount of memory space. But if the flow rate of the data needs to be processed, the processing power of the integrated classifier is exceeded. Integrated classifiers can't train all the date which have arrived recently. According to this, presents an improved reservoir sampling algorithm. The algorithm and C4.5 decision tree classification algorithm for coupling, design the integrated classifier algorithm improved high speed data stream based on the reservoir sampling, the algorithm for the goal of the algorithm, is a sub linear space algorithm, can effectively shorten the training time in updating the integrated classifier at the same time, the classification performance of classifier remains high. In addition, the use of weighted random sampling algorithm and C4.5 decision tree classification algorithm for coupling, the design of the control algorithm, compared with the target hyper-plane algorithm in the artificial data set, to verify the superiority of the algorithm.
机译：数据流挖掘需要在占用少量存储空间的情况下快速处理数据并进行自适应概念漂移。但是，如果需要处理数据流率，则会超出集成分类器的处理能力。集成分类器无法训练最近到达的所有日期。据此，提出了一种改进的油藏采样算法。该算法与C4.5决策树分类算法进行耦合，设计综合分类器算法，改进了基于油藏采样的高速数据流，该算法针对该算法，是一种亚线性空间算法，可以有效缩短训练时间在更新集成分类器的同时，分类器的分类性能仍然很高。另外，利用加权随机抽样算法和C4.5决策树分类算法进行耦合，设计了控制算法，并与目标超平面算法在人工数据集中进行了比较，验证了该算法的优越性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号