| 基于支持向量机算法和其他算法在文本分类中的性能比较 |
| |
| 引用本文： | 木林.基于支持向量机算法和其他算法在文本分类中的性能比较[J].内蒙古大学学报(自然科学版),2011,42(6):703-707. |
| |
| 作者姓名： | 木林 |
| |
| 作者单位： | 呼伦贝尔学院计算机科学与技术学院,内蒙古呼伦贝尔,021008 |
| |
| 基金项目： | 内蒙古自治区高等学校科学研究项目(NJZZ11230) |
| |
| 摘 要： | 分析了几种典型的文本分类算法的特点.其中,支持向量机算法具有最优性能,但处理大量数据时,耗时比较大;K-NN方法与K值的选取有很大关系,而且实时性不够良好;朴素贝叶斯方法在实际应用中可以得到比较好的效果;神经网络方法提供了比较容易的方式预测非线性系统,训练过程很慢,不能适应大数据量的学习.
|
| 关 键 词： | 支持向量机方法 K-NN方法 朴素贝叶斯方法 神经网络方法 |
Performance Comparison Based on Support Vector Machine Algorithm and Other Algorithms in Text Categorization |
| |
| MU Lin.Performance Comparison Based on Support Vector Machine Algorithm and Other Algorithms in Text Categorization[J].Acta Scientiarum Naturalium Universitatis Neimongol,2011,42(6):703-707. |
| |
| Authors: | MU Lin |
| |
| Institution: | MU Lin(School of Computer Science and Technology,Hulunbeir College,Hulunbeir 021008,China) |
| |
| Abstract: | Characteristics of several typical text categorization algorithms are analyzed.Among these algorithms,the support vector machine algorithm has the optimal performance,but when large amounts of data to be processed,it takes a long time;K-NN method has much to do with the selection of K value,and its real-time performance is not good enough;Naive Bayes method can get better results in practice;Neural network method provides a relatively easy way to predict nonlinear systems,but the training process is very sl... |
| |
| Keywords: | support vector machine method K-NN method Naive Bayes method neural network method |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
