News text clustering optimization based on improved DPC algorithm

In news text clustering optimization, the current text clustering optimization methods have the drawback of poor clustering optimization effect. In response to this issue, this study introduces an improved density peak clustering algorithm to cluster news text data, and uses the k-means clustering algorithm to optimize mixed text. On the basis of improving the density peak clustering algorithm, a new clustering optimization method is proposed by combining the feature word pairing method that can extract features from text information. The experiment showed that the research algorithm had an average absolute error and root mean square error of 1.035 and 0.963 when optimizing the clustering of news texts, and the clustering accuracy of this algorithm reached 94.36%, significantly higher than other algorithms. The optimization method achieved an accuracy of 89.67% in extracting text feature words for clustering optimization of different types of news texts, which was significantly higher than the comparison method. The above results indicate that the proposed method has a good clustering optimization effect on news texts, providing a technical support for the field of text clustering research.