| 网络舆论话题分类算法研究与实证分析 |
| |
| 引用本文： | 柴丹炜,邵思思,张若昕,乐光学,刘建生. 网络舆论话题分类算法研究与实证分析[J]. 嘉兴学院学报, 2014, 26(6): 116-123 |
| |
| 作者姓名： | 柴丹炜 邵思思 张若昕 乐光学 刘建生 |
| |
| 作者单位： | 1. 嘉兴学院数理与信息工程学院,浙江嘉兴,314001
2. 江西理工大学理学院,江西赣州,341000 |
| |
| 基金项目： | 嘉兴学院大学生研究训练(SRT)计划项目 |
| |
| 摘 要： | 针对网络舆论话题的复杂性,设计实现了一种基于主成分分析法和朴素贝叶斯的网络舆论话题分类算法。首先使用主成分分析法对建立的舆论话题内容向量进行降维处理,并去除列向量之间的相关性,然后使用朴素贝叶斯算法进行分类分析,最后用实验对算法的有效性进行了验证。通过与其他分类算法的对比,可得到基于主成分分析法和朴素贝叶斯的网络舆论话题分类算法的分类效果或算法开销要优于其他一些分类算法.
|
| 关 键 词： | 舆论话题 文本分类 主成分分析 朴素贝叶斯 |
Research and Empirical Analyses of the Classification Algorithm of Network Public Opinion Topics |
| |
| CHAI Danwei,SHAO Sisi,ZHANG Ruoxin,YUE Guangxue,LIU Jiansheng. Research and Empirical Analyses of the Classification Algorithm of Network Public Opinion Topics[J]. Journal of Jiaxing College, 2014, 26(6): 116-123 |
| |
| Authors: | CHAI Danwei SHAO Sisi ZHANG Ruoxin YUE Guangxue LIU Jiansheng |
| |
| Affiliation: | CHAI Danwei, SHAO Sisi, ZHANG Ruoxin, YUE Guangxue, LIU Jiansheng( 1. School of Mathematics Physics and Information Engineering, Jiaxing University, Jiaxing, Zhejiang 314001 ; 2. College of Science,Jiangxi University of Science and Technology,Ganzhou,Jiangxi 341000) |
| |
| Abstract: | Due to the complexity of network public opinion topics,a classification algorithm for the network topics is designed and implemented,based on principal component analysis and Naive Bayes.The principal component analysis is firstly used to reduce the dimension of the component vectors and remove the correlation between each column vector.Then the Naive Bayes algorithm is used for classification analysis.Finally,the validity of the algorithm is verified by some experiments.According to the comparison with other classification algorithms,we conclude that the classification algorithm based on the principal component analysis and Naive Bayes does better than other classification algorithms in the classification effect and the algorithm cost. |
| |
| Keywords: | public opinion topic text classification principal component analysis Naive Bayes |
| 本文献已被 维普 等数据库收录！ |
|
