论文部分内容阅读
目标多属性序列类数据不能直接与数据库中的区间类数据融合识别,对此,提出一种基于云变换的序列-区间异类数据识别算法.对目标多属性序列数据进行频数计算形成频率分布函数,并进行虚警检测,实施云变换形成云簇,提取云簇特征,再根据3 En准则形成云滴区间,实现了序列型数据的区间化表示.进一步,利用一种区间多属性识别判定准则进行识别判定,得到识别结果,解决了序列-区间异类数据的识别问题.仿真实验结果验证了该算法对序列-区间异类数据识别的有效性.
To solve this problem, we propose a sequence-interval heterogeneous data recognition algorithm based on cloud transformation.We calculate the frequency of the target multi-attribute sequence data to form the frequency distribution function, And detect the false alarm, perform cloud transform to form cloud cluster, extract cloud cluster characteristics, and then form cloud drop interval according to 3 En criterion, and achieve the intervalized representation of sequence data.Furthermore, an interval multi-attribute identification criterion Recognition and recognition, the recognition of sequence-interval heterogeneous data is solved.The simulation results verify the effectiveness of the algorithm in identifying sequence-interval heterogeneous data.