• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议> >New Measure of Interestingness for Efficient Extraction of Association Rules
【24h】

New Measure of Interestingness for Efficient Extraction of Association Rules

机译：有效提取关联规则的兴趣度的新度量

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Data Mining helps to uncover the already unknown and non-redundant knowledge in large databases, which can be used for decision making purpose. Association rule mining is one of the key research area in the field of Data Mining. Association rule mining can be considered as unsupervised learning model, it discovers the interesting relationship among large set of data items on the basis of some predefined threshold. Support-confidence is the classical model used for the rule mining purpose, it uses confidence for final rule generation but it has some limitations. As sometimes it can generate those rules which are not positively correlated and thus can mislead the decision maker. In this paper we addressed the problems associated with existing approach and also proposed two new measure of interestingness to deal with these problems. The new measures have been tested for their correctness.
机译：数据挖掘有助于发现大型数据库中已经未知和非冗余的知识，这些知识可用于决策目的。关联规则挖掘是数据挖掘领域的关键研究领域之一。关联规则挖掘可以看作是无监督的学习模型，它基于一些预定义的阈值发现大量数据项之间的有趣关系。支持置信度是用于规则挖掘目的的经典模型，它使用置信度来生成最终规则，但有一些局限性。有时它会生成那些没有正相关的规则，从而可能误导决策者。在本文中，我们解决了与现有方法相关的问题，并提出了两种有趣的新方法来解决这些问题。新措施已经过正确性测试。

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