| 基于K-近邻方法的网络信息文本分类 |
| |
| 引用本文： | 刘开袆,江志雄.基于K-近邻方法的网络信息文本分类[J].贵州大学学报(自然科学版),2009,26(3):60-63. |
| |
| 作者姓名： | 刘开袆 江志雄 |
| |
| 作者单位： | 1. 贵州大方发电有限公司,贵州,毕节,551600
2. 上海海港通信技术有限公司,上海,200043 |
| |
| 摘 要： | 提出了一种网络信息文本分类模型的建立方法。根据网络报文的特点，抽取其中关键词作为分类特征词条，并以报文关键词进行词频统计分析建立文本分模型。分别进行了基于最近邻决策和K-邻近决策的分类效果试验研究，结果显示：K-近邻决策的分类效果要优于最近邻决策的分类效果。
|
| 关 键 词： | 文本分类 最近邻 K-近邻 特征提取 |
Network Text Classification Based on K-Nearnest neighbor Method |
| |
| LIU Kai-yi,JIANG Zhi-Xiong.Network Text Classification Based on K-Nearnest neighbor Method[J].Journal of Guizhou University(Natural Science),2009,26(3):60-63. |
| |
| Authors: | LIU Kai-yi JIANG Zhi-Xiong |
| |
| Institution: | 1.Guizhou Dafang Power Generation Co.Ltd;Bijie 551600;China;2.Shanghai Haigang Communication Technology Corporation;Shanghai 200043;China |
| |
| Abstract: | One network information text categorization construction model was proposed.According to the feature of network datagram,the keywords were selected as classification features and computer the keywords frequency to construct a text categorization model.The experiments show that the K-nearest neighbor method is superior to the nearest neighbor method. |
| |
| Keywords: | text categorization nearest neighbor k-nearest neighbor feature selection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
