An algorithm for rationality estimation of movement trend based on big data analysis

Abstract

The traditional reasonable estimation algorithm of data movement trend has some problems, such as easily falling into local optimum, high computational complexity, and poor generality. This paper presents a reasonable estimation algorithm of data movement trend based on big data analysis. The combination of particle swarm optimization (PSO) and mutation operation can prevent the stagnation of particle swarm evolution and improve the estimation ability. While using the PSO algorithm for global search, it combines the characteristics of the movement vector effectively, selects the appropriate particle population, and uses the appropriate termination strategy to reduce the computational complexity and obtain the constraint conditions of the trend trajectory of the data movement. Through the database system center server and the location server, the data access is carried out co processing and querying, and the differential evolution algorithm is used to estimate the original database in the system initialization stage. Experiments show that the proposed algorithm can reasonably estimate the trend of data movement under big data analysis, with low computational complexity and good versatility.