| 中文网络评论的复杂语义倾向性计算方法研究 |
| |
| 引用本文： | 郝玫,王道平.中文网络评论的复杂语义倾向性计算方法研究[J].图书情报工作,2014,58(22):105. |
| |
| 作者姓名： | 郝玫 王道平 |
| |
| 作者单位： | 北京科技大学东凌经济管理学院 |
| |
| 基金项目： | 本文系中央高校基本科研业务费项目“敏捷供应链中的客户评价倾向和产品推荐研究”(项目编号:FRF-BR-14-012A)研究成果之一. |
| |
| 摘 要： | 考虑到中文网络评论的复杂语义特性,为提高产品特征观点倾向性判断的精确性,提出一种复杂语义倾向性计算方法.该方法在建立产品领域情感词典的基础上,首先确定特征观点窗口的度量范围,完成特征观点组的提取;然后在特征观点组中综合考虑观点词的程度、反转语义及特征评价的频数等多种因素,完成特征评价倾向性的计算.实验结果表明,本文所提出的方法在特征评价倾向性分类方面可以达到较高的查全率和查准率,而且与SO-PMI方法相比,可明显提高特征评价的计算精确性.
|
| 关 键 词： | 中文网络评论 特征评价 情感倾向 复杂语义 |
| 收稿时间： | 2014-09-30 |
| 修稿时间： | 2014-11-02 |
Complex Semantic Orientation of Chinese Online Customer Reviews |
| |
| Hao Mei,Wang Daoping.Complex Semantic Orientation of Chinese Online Customer Reviews[J].Library and Information Service,2014,58(22):105. |
| |
| Authors: | Hao Mei Wang Daoping |
| |
| Institution: | Dongling School of Economics and Management, University of Science and Technology Beijing, Beijing 100083 |
| |
| Abstract: | Considering the complex semantic of Chinese online reviews, this paper proposes a complex semantic orientation calculation method to improve the evaluation accuracy of product features. Based on building the product sentimental dictionary, the method determines the size of feature opinion window and extracts the feature opinion set, and then it calculates the feature evaluation combined with opinion degree, semantic inversion and features evaluation frequency. The experimental results show that the proposed method can achieve higher recall ratio and precision in terms of feature evaluation classification, and compared with SO-PMI, it can obviously improve the calculation accuracy of feature evaluation. |
| |
| Keywords: | Chinese online reviews feature evaluation emotional orientation complex semantic |
|
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
|
