重复投影数据库下的优化挖掘方法研究与仿真

Study and Simulation of Optimized Mining Method under Repeat Projection Database

进行重复投影会使数据库中出现大量冗余数据,使得数据特征之间出现相似性干扰。传统的挖掘方法,遇到特征相似性情况时,以模糊规则建立挖掘关联规则,在建立模糊规则时,一旦特征过于接近,需要加入很多的约束条件,计算过程较为繁琐,导致数据挖掘准确性差、效率低的问题。提出基于贝努里模型的大量重复投影数据库下优化挖掘方法,依据等级分组识别重复投影数据,建立贝努里模型。根据冗余属性的贡献及区分能力的不同,确定两个分类贡献系数。依据两个先验概率值,依次放大两个分类贡献系数倍。通过改进贝努里事件模型中条件概率中的估计方法,划分大量重复投影数据库下的重复数据,实现大量重复投影数据库下的优化挖掘,提高数据搜索效率。仿真结果表明,改进方法不仅具有较优的挖掘性能,而且挖掘效率高。

An optimized mining method based on the Bernoulli model under a large number of repeat projection database is proposed.According to the grade grouping,repeat projection data are identified,and the Bernoulli model,is established.According to the contribution differences of the redundant attributes and the separating capacity,two classification contribution coefficients are determined.Based on the two prior probability values,the two classification contribution coefficients are made magnification in turn.By improving the estimation method of conditional probability in Bernoulli event model,a large number of duplicate data under repeat projection database are divided,the optimization mining under a large number of repeat projection database is realized,and the efficiency of data search is improved.The simulation results show that the improved method not only has good performance with high mining efficiency.