基于最小二乘支持向量机的网页主题语义分类的研究

ON SEMANTIC CLASSIFICATION OF WEBPAGE SUBJECTS BASED ON LEAST SQUARES SUPPORT VECTOR MACHINES

提出了对网页主题进行语义扩展的方法,利用最小二乘支持向量机LSSVM(least squares support vector machines)来代替传统的支持向量机SVM(support vector machine)的分类技术。在建立LSSVM模型的多类别分类算法基础上,将其应用到网页主题语义分类。实验表明,最小二乘支持向量机学习速度快,在小样本情况下具有良好的非线性建模和泛化能力,对网页主题语义分类具有很好的效果。

This paper proposes a new method of semantic extension for webpage subjects by utilizing the least squares support vector machines （LSSVM） to replace the classification technique with traditional support vector machine （SVM）. On the basis of setting up the multiclass classification algorithm of LSSVM model, the authors applied the method to semantic classification of webpage subjects. The experimental results show that LSSVM model has some advantages such as fast learning speed, nonlinearity modelling and generalization power in condition of small number of examples, and has good performance on semantic classification of webpage subjects.