188宝金博页面版

  • 图案背景
  • 纯色背景
视图
标记
批注
批注本地保存成功，开通会员云端永久保存 去开通
nqctn

上传于：2018-03-30

粉丝量：15

该文档贡献者很忙，什么也没留下。


  • 相关
  • 目录
  • 笔记
  • 书签

188宝金博页面版:更多相关文档

  • A METHOD OF CLASSIFICATION AND ASSOCIATION RULES OBTAINING

    星级： 79 页

  • [精品]Meaningful discretization of continuous features for association rules mining by means of a

    星级： 6 页

  • finding interesting rules from large sets of discovered association rules

    星级： 32 页

  • finding interesting rules from large sets of discovered association rules

    星级： 11 页

  • a method of classification and association rules obtaining

    星级： 103 页

  • Effective Diagnosis of Alzheimer’s Disease by Means of Association Rules

    星级： 8 页

  • Classification Using Association Rules - IIMA

    星级： 52 页

  • Incremental Mining of Association Rules

    星级： 9 页

  • MIDClass Microarray Data Classification by Association Rules and Gene Expression Intervals

    星级： 8 页

暂无目录

点击鼠标右键菜单,创建目录

    暂无笔记

    选择文本，点击鼠标右键菜单，添加笔记

    暂无书签

    在左侧文档中，点击鼠标右键，添加书签

    188宝金博页面版: Majority Classification by Means of Association Rules

    下载积分： 9950

    内容提示： Majority Classif i cationby Means of Association RulesElena Baralis and Paolo GarzaPolitecnico di TorinoCorso Duca degli Abruzzi 24, 10129 Torino, Italy{baralis,garza}@polito.itAbstract. Associative classif i cation is a well-known technique for struc-tured data classif i cation. Most previous work on associative classif i cationbased the assignment of the class label on a single classif i cation rule. Inthis work we propose the assignment of the class label based on simplemajority voting among a group of ...

    文档格式：PDF | 页数：12 | 浏览次数：62 | 上传日期：2018-03-30 07:38:56 | 文档星级：
    Majority Classif i cationby Means of Association RulesElena Baralis and Paolo GarzaPolitecnico di TorinoCorso Duca degli Abruzzi 24, 10129 Torino, Italy{baralis,garza}@polito.itAbstract. Associative classif i cation is a well-known technique for struc-tured data classif i cation. Most previous work on associative classif i cationbased the assignment of the class label on a single classif i cation rule. Inthis work we propose the assignment of the class label based on simplemajority voting among a group of rules matching the test case.We propose a new algorithm,L 3M , which is based on previously proposedalgorithm L 3 . L 3 performed a reduced amount of pruning, coupled witha two step classif i cation process. L 3M combines this approach with the useof multiple rules for data classif i cation. The use of multiple rules, bothduring database coverage and classif i cation, yields an improved accuracy.1 IntroductionAssociation rules [1] describe the co-occurrence among data items in a largeamount of collected data. Recently, association rules have been also considereda valuable tool for classif i cation purposes. Classif i cation rule mining is the dis-covery of a rule set in the training database to form a model of the data, theclassif i er. The classif i er is then used to classify appropriately new data for whichthe class label is unknown [12]. Dif f erently from decision trees, association rulesconsider the simultaneous correspondence of values of dif f erent attributes, henceallowing to achieve better accuracy [2,4,8,9,14].Most recent approaches to associative classif i cation (e.g., CAEP [4], CBA[9], ADT [14], and L 3 [2]) use a single classif i cation rule to assign the class labelto new data whose label is unknown. A dif f erent approach, based on the use ofmultiple association rules to perform classif i cation of new data has been proposedin CMAR [8], where it has been shown that this technique yields an increasein the accuracy of the classif i er. We believe that this technique can be appliedorthogonally to almost any type of classif i er. Hence, in this paper we proposeL 3M , a new algorithm which incorporates multiple rule classif i cation into L3 , alevelwise classif i er previously proposed in [2].L 3 was based on the observation that most previous approaches, when per-forming pruning to reduce the size of the rule base obtained from associationrule mining, may go too far and discard also useful knowledge. We extend thisN. Lavraˇ c et al. (Eds.): PKDD 2003, LNAI 2838, pp. 35–46, 2003.c ? Springer-Verlag Berlin Heidelberg 2003

    阅读了该文档的用户还阅读了这些文档

    188宝金博页面版:关于我们

    188宝金博页面版:帮助中心

    188宝金博页面版:关注我们

    • 新浪微博

    关注188宝金博页面版公众号

    188宝金博页面版
    阅读
    APP
    阅读
    返回
    顶部
    188宝金博页面版官网登录在线平台入口（2026已更新）—江苏协昌电子科技股份有限公司