Support Vector Regression Prediction Method of Text Data Based on Correlation Analysis

Correlation analysis is an important research content of text data prediction. Experiments have found that there are continuous distribution, delay mutation and mutation amplitude distribution characteristics in sequence delay correlation analysis. It is proved that when the amount of text data is large, support vector regression is used. The prediction method can achieve smaller errors and even solve the problem of delay mutation. According to experimental phenomena, the proposed method can overcome the shortcomings of existing algorithms and effectively handle most of the delay mutations in the process of text data prediction. A large number of experiments have proved that this paper can achieve better performance than existing methods.