| 基于并行决策树的微博互动数预测 |
| |
| 引用本文： | 黄林昊,郭昆.基于并行决策树的微博互动数预测[J].福建工程学院学报,2017,0(3):294-300. |
| |
| 作者姓名： | 黄林昊 郭昆 |
| |
| 作者单位： | 福建广播电视大学电子信息与计算机系 |
| |
| 摘 要： | 社交网络的快速发展，微博成为主要的社交媒体平台，针对如何预测微博文本的未来互动数，对微博进行有效的分发控制的问题，提出一种基于并行决策树的微博互动数所属级数预测的方法。首先，对用户以往发表的微博进行用户特征和微博文本特征的处理；然后，使用并行决策树分类算法对训练数据进行分类模型的构建；最后使用得到的分类模型对新微博文本的互动数所属级数进行分类预测。通过对比算法的实验，验证了所提方法具有较高的分类精度和较好的可扩展性，能够对微博所属级数进行有效的分类预测。
|
| 关 键 词： | 微博 互动数 并行 决策树 预测 |
Interaction number prediction of micro-blog based on parallel decision tree |
| |
| Huang Linhao,Guo Kun.Interaction number prediction of micro-blog based on parallel decision tree[J].Journal of Fujian University of Technology,2017,0(3):294-300. |
| |
| Authors: | Huang Linhao Guo Kun |
| |
| Institution: | Electronic Information and Computer Department, Fujian Radio and TV University |
| |
| Abstract: | To predict the future interaction number of micro-blog texts to implement effective distribution control of micro-blogs, a method of forecasting the series number of micro-blog interaction numbers based on parallel decision tree was proposed. Firstly, the user characteristics and micro-blog text features of the user’s previous micro-blog were processed. Then, a classification model of the training data was constructed via a parallel decision tree classification algorithm. Finally, the series number of the interaction number of new micro-blog texts was classified via the classification model. The experimental results show that the proposed method has high classification accuracy and good scalability and can effectively forecast micro-blog series. |
| |
| Keywords: | micro-blog interaction number parallel decision tree forecast |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《福建工程学院学报》浏览原始摘要信息 |
| 点击此处可从《福建工程学院学报》下载免费的PDF全文 |
