A new SVM Chinese text of classification algorithm based on the semantic kernel

Popular Chinese text classification algorithms are mostly based on word frequency statistics features, ignoring the characteristics of Chinese text between the semantic relevance. To further improve the Chinese text classification results, the paper presents a new semantic-based kernel of SVM algorithm for Chinese text classification, through simple idea and smaller implementation costs. Experiments show that compared with traditional SVM algorithm, the algorithm in the Chinese text classification efficiency and accuracy has significantly improved, with good classification results.