一种改进的Apriori算法在移动图书馆中的应用

摘要

科技的高速发展使移动图书馆已经成为阅读的新趋势，而数据挖掘技术作为一种信息资源智能化的前沿技术，为移动图书馆的人性化需求提供了重要的支持和保障。用传统的Apriori算法对移动图书馆数据进行挖掘，可以找出常规的频繁项集和关联规则，但存在运算效率不高、极具价值但出现频率相对较低的项集易被忽略等不足。为了解决这种缺陷，提出了一种改进的Apriori算法，该算法通过关联数据与原始数据的关联程度赋予权值，计算出其加权支持度与加权置信度而得到关联规则。通过实例对该算法进行验证。结果表明，该算法可以挖掘出传统的Apriori算法得到的关联规则，同时还可以挖掘出频率不高但极具价值的新关联规则，使挖掘结果更具价值。%An improved Apriori algorithm is presented in this paper.It gives weight values based on the association de-gree of correlation data and original data,calculates the weighted support degree and confidence degree,and then achieves the desired association rules. Final y,the algorithm is verified by experiments. The results show that the improved algorithm can find association rules in perfect accordance with the traditional Apriori algorithm,but also can dig out the new associa-tion rules which are very valuable but can't be found by the traditional one due to the lower frequency of occurrence.