【摘要】 针对常用的文本分类算法,给定五种文本类型的数据集,通过使用典型的文本分类算法进行实验分析,通过精确率、召回率和测试值的精度来评估这些文本分类器的性能,并给出分析结果和改进的组合训练方法。结果表明:将半监督学习训练和监督学习相结合能达到更好的分类效果。为了提高文本推荐速度,前期工作就是要选择合适的分类算法方法,组合选择算法,提高准确度和效率。更多还原

【Abstract】 Analyzes several typical text classification algorithms, gives five types of text data sets, the classic text categorization algorithm test com-parison by precision, recall accuracy rate and test value to evaluate the performance of the text classifier, and gives the analysis result and the improved combination training method. The results show that the combination of semi supervised learning training and supervised learning can achieve better classification results. In order to improve the speed of text recommendation, the preliminary work is to choose the appropriate classification algorithm, combine selection algorithm to improve the accuracy and efficiency.更多还原