Incremental Attribute Reduction Algorithm for Smart City Local Area Communication Systems Based on Similarity Relation

Abstract

Abstract The wide application of local area communication system brings comprehensive data information but also increases the difficulty of data mining and analysis. Therefore, the data mining preprocessing link-attribute reduction is studied. The research is divided into three parts: first, the method of distinguished matrix fast calculating is used for discerning the core attributes of a data set; second, the k-nearest neighbor algorithm is used to calculate the attribute as well as the similarity coefficient between condition attributes, and to finish at the beginning of attribute reduction; and third, the global optimization ability of particle swarm algorithm implementation attribute reduction is used again to complete local area communication system incremental attribute reduction targets. The results show that compared with the three previous reduction algorithms, the proposed algorithm has the least number of attributes and the least number of iterations, which proves the reduction degree and efficiency of the proposed method.