首页    期刊浏览 2024年12月01日 星期日
登录注册

文章基本信息

  • 标题:Exposition and Execution of a Scaled Aprori in Multidimensional Dataset
  • 本地全文:下载
  • 作者:M Afshar Alam ; Sapna Jain ; Ranjit Biswas
  • 期刊名称:ARPN Journal of Systems and Software
  • 电子版ISSN:2222-9833
  • 出版年度:2012
  • 卷号:2
  • 期号:1
  • 页码:27-36
  • 出版社:ARPN Publishers
  • 摘要:

    This research work proposes discovery of quantitative association rule in spatial data that is related only to objects what enables data to be stored in relational database management system .The traditional Apriori algorithm is a method that helps to minimize the number of candidate sets while generating association rules by evaluating quantitative information associated with each item .We have proposed an algorithm for Aprori-UB which uses multidimensional access method, UB-tree to generate better asso-ciation rules with high support and confidence. In multidimensional databases, objects are indexed according to several or many independent attributes. However, this task cannot be effectively realized using many standalone indices and thus special indexing structures have been developed is last two decades. Common to all this structures is that they index vectors of values instead of indexing single values. The UB-tree represents one of the promising multidimensional index structures. Indexing and querying high-dimensional databases is a challenge for current research since high-dimensional indexing is significantly influenced phenomenon called curse of dimensionality .The proposed method in the paper reduces the number of item sets generated and also improves the execution time of the algorithm. Any valued attribute will be treated as quantitative and will be used to derive the quantitative association rule which usually helps in making the rules efficient to handle all types of data.

  • 关键词:Data Mining; Association Rule; Quantitative multidimensional theorem; Apriori algorithm; Frequent Itemset; UB-tree
国家哲学社会科学文献中心版权所有