【摘要】 为方便中文文本分类过程算法研究,阐述中文文本分类的基础技术,提出基于构件的中文文本分类技术研究辅助平台。该平台对大多数分类中使用的算法在中文文本分类中的应用效果进行了研究。实验结果表明:该平台可以通过计算分类器分类准确率的宏平均值比较分类算法与特征选择算法的性能,可以评估语料库的可用性,能够用于研究中文分词、特征选择、分类算法等中文文本分类技术问题。更多还原

【Abstract】 This paper presents a support platform for Chinese text categorization based on components.Through this platform we studied the performance of most of traditional algorithms when they are now used in Chinese text categorization. The results show that the platform can be used to compare the performance of various classification algorithms and feature selection algorithms by analyzing the average classification accuracy of classifiers and to assess the availability of the corpus.The platform can also be used to study text categorization technologies including Chinese word segmentation,feature selection and classification algorithms.更多还原