【摘要】 关联规则挖掘可以发现大量数据中项集之间有趣的关联或相关联系,并已在许多领域得到了广泛的应用。目前业界已经提出了许多发现关联规则的算法,这些算法都认为每个数据对规则的重要性相同。但在实际应用中,用户会比较倾向于自己最感兴趣或认为最重要的那部分项目,因此有必要加强这些项目对规则的影响,同时减弱另一些用户兴趣不大或认为不重要的项目对规则的影响。为此,论文提出了水平加权关联规则的问题,并结合Apriori算法,加以改进,给出了关于该问题的解决方案及有效算法New_Apriori。更多还原

【Abstract】 Mining association rules can find out some interesting associations or correlations between items among large quantity of data and has many wide applications in some fields.Now,lots of algorithms ha ve been proposed for finding the association rules.Most of these algorithms tr eat each item as uniformity.However,in real applications,users are more incli ned to such items as they are most interested in or feel most important about .So,it’s necessary to emphasize the affection of such items and reduce the af fection of other items to rules.Therefore,this paper proposes the problem of mining horizontal weighted association rules,advances the algorithm Apriori,an d presentes an effective algorithm New_Apriori to handle the problem of mining horizontal weighted association rules.更多还原