Attribute Reduction Method of Power Big Data Based on Rough Set

Based on the consistency between condition attributes and decision labels, classical rough set attribute reduction can effectively remove redundant attributes from symbolic data. To deal with different types and even mixed high-dimensional data, the classical rough set is extended to a more generalized model. However, in the face of large-scale and dynamic data, the existing rough set attribute reduction algorithms often consume a lot of computing time, and even fail to execute in some hardware and software environments due to memory overflow. To overcome these problems, aiming at the problem of high reduction rate in traditional attribute reduction methods of power big data, an attribute reduction method of power big data based on a rough set is proposed. The method comprises the following steps of: performing granulation processing on the electric power big data according to neighborhood measurement, acquiring the neighborhoods of all objects in a data real number space, sorting the electric power big data attributes by using a quick sorting algorithm, and performing discretization on an electric power large data attribute sequence according to a rough set theory to reduce the data attributes. The experimental results show that the attribute reduction rate of the design method is higher than that of the traditional method, and it has good feasibility and reliability.