Research on text feature clustering based on improved parallel genetic algorithm

Synonyms, synonyms and the strong association of semantic information increase the dimension text feature vectors, and greatly affect the efficiency and accuracy of text classification. In order to reduce the dimension of the text feature vectors, this paper presents an improved parallel genetic algorithm to sovle the text feature clustering problem. Firstly, the K-means algorithm is used to cluster the feature words. The parallel genetic algorithm is used to fine-grained the feature words. In the process of applying genetic algorithms, crossover operator is improved so that the algorithm has a global search ability and local search capability, and reduce the dependence on the initial cluster centers. Finally, the different types of feature words are analyzed and compressed, and a set of feature words of the text category and the semantic information is formed. The experimental results verify our method of text feature extraction is better.

I. Introduction

Feature words clustering is a way by clustering a set of word processing text feature. The feature words of text are divided into several clusters, which makes the same cluster in the feature words with the greatest possible degree of similarity, while keeping the inter-cluster feature words similarity as small as possible. In the field of natural language processing, Feature Words Clustering is often used, such as feature extraction, text categorization and semantic association analysis. It received wide attention of scholars at home and abroad due to its importance [1] [2].