The Research of Chinese Text Automatic Classification Based on Multiple

Abstract

Support Vector Machines (SVMs) is a new technique for data mining. It has wide applications in various fields and is a research hot pot of the machine learning field, but, being applied to handling large-scale problems, SVMs needs longer t raining time and larger memory. It’s an effective way to solve large scale data processing in text classification with multiple classifier systems composed by multiple support vector machine classifiers. Based on the analysis of traditional parallel algorithms, this paper proposes an improved algorithm based on multiple SVMs. The experimental results indicate that the new algorithm works well in precision and recall rate in the condition that the speeds of classification increase remarkably. Compared with traditional algorithms, the classified accuracy is lower but is within the range for acceptance.