【摘要】 构建精确而有效的分类器是数据挖掘和机器学习中的一个重要任务.提出了一种基于频繁模式树的关联分类规则挖掘算法,该算法同时考虑所有属性,并对现有关联分类规则挖掘算法中内存要求高、类别属性处理难、I/O访问次数多等问题提出了相应的解决方案.试验结果表明,该方法可以取得比同样基于关联规则的分类算法CMAR更高的执行效率以及基于规则的决策树分类算法C4.5更好的分类效果.更多还原

【Abstract】 Building accurate and efficient classifier for large databases is an essential task of data mining and machine learning research.An algorithm for mining associative classification rules based on frequent pattern tree is presented,which considers all attributes at one time,and helps to solve a number of pro-(blems) that exist in the current classification systems,such as high memory request,difficult catalog attri-butes management,repetitious I/O accessing,and so on.The experiments show that the method has better classification accuracy in comparison with C4.5.Moreover,the performance study shows that the method is highly efficient in comparison with other reported associative classification(methods.)更多还原