Operation data prediction algorithm of information system based on discrete second-order difference

Beining Sun

Abstract: This study proposes an algorithm for predicting the running data of information systems based on discrete second-order difference clustering. The wide stationary time series model of information system operation data is established, and the association rules mining and feature distributed transmission sequence fitting of information system operation data are conducted by binary semantic information representation method. The principal component feature detection and matching of information system operation data are carried out. High-order spectral feature analysis and extraction of information system operation data is realized based on big data analysis, and the prediction algorithm is improved. The proposed method has high accuracy, good convergence and high real-time performance, which can improve the scheduling ability of information system operation data.