Research on Intelligent Clustering Scoring of English Text Based on XGBOOST Algorithm

In today's explosive growth of online text data, how to feature mine text data from a wide range of sources, cluster text data with similar features, and classify them according to their features has become a hot issue. In this paper, according to the characteristics of English text data, considering the existence of a large number of redundant features in English text, the existence of features with similar meanings in English text and other problems, the following method is proposed: in the first step, the English text is segmented, data preprocessing is carried out, and then the text data is processed by using the TF-IDF algorithm, to get the roughly selected subset of features; in the second step, the improved Binary Particle Swarm Optimization algorithm is used to In the second step, the improved binary particle swarm optimization algorithm GSBPSO (Global-to-local Searching-based BinaryParticle Swarm Optimization) is used to re-select the features of the roughly selected feature subset obtained in the first step to obtain the optimal feature subset; in the third step, the binary SKM clustering algorithm is used to cluster English text according to the optimal feature subset obtained in the second step. English text is clustered according to the optimal feature subset obtained in the second step, and the classification criteria are formulated for the existing text corpus to categorize the text data. Finally, the XGBOOST algorithm is applied to evaluate the prediction and establish the intelligent clustering scoring model of English text based on XGBOOST.