基于维度变化的矩阵增量属性约简算法

Matrix incremental attribute reduction algorithm based on dimension change

随着计算机网络技术和人们生活节奏的加快,生活中很多数据都在随时发生着变化,那么快速及时的解决数据变化后的属性约简问题,就成了信息技术领域里研究的一个重要课题。剖析了数据更新后相对知识粒度和等价关系矩阵的增量机制,提出了对象属性值增加后的基于矩阵方法的增量属性约简算法。下载了2组UCI数据对提出的增量属性约简算法进行了测试,结果证明了增量属性约简算法能够处理属性值增加后的属性约简问题。

With the acceleration of computer network technology and people’s pace of life,a lot of data in life are changing at any time.Quickly and timely solving the problem of attribute reduction after data changes has become an important topic in the field of information technology research.In this paper,the incremental mechanism of relative knowledge granularity and equivalence relation matrix after data update is analyzed.Then an incremental attribute reduction algorithm is proposed,which is after object attribute value increasing and based on matrix method.Finally,two groups of UCI data are downloaded to test the algorithm,and the results show that the incremental attribute reduction algorithm can deal with the attribute reduction problem with increased attribute values.