纽块分析是一种非常重要的句法分析预处理手段,通过将文本划分成一组互不重叠的片断,来达到降低句法分析的难度。提出一种基于SVM—Adaboost的中文纽块分析方法,将基于线性核函数的支持向量机与Adaboost算法相结合,以基于线性核函数的SVM作为Adaboost的分量分类器,在学习过程中改变分量分类器的核参数。实验结果表明了该算法的有效性。
Text chunking is a very important approach to preprocessing parsing.It divides text into syntactically related non-overlapping groups of chunks in order to reduce the complexity of the full parsing.In this paper,a SVM-Adaboost algorithm is applied for Chinese text chunking which combines Adaboost with linear-kernel SVM.This algorithm uses SVM as weak learners for AdaBoost and adjusts the kernel parameter of SVM in the learning process.The experimental results show that it is an effective approach.