Affiliation of Author(s):信息科学技术学院
Journal:Application and techniques in information security
Funded by:国家自然科学基金项目
Abstract:Mining opinions and sentiment from cross-cultural communication Web sites can deepen mutual understanding among people between countries and provides an important channel for researching China’s cross-cultural communication. The sentiment analysis in the context of cross-cultural communication faces the challenges of culture-dependent, ne-grained sentiment understanding, and topic-centralization. Traditional approaches use machine learning methods, such as Naive Bayes, maximum entropy and support vector machine. In this paper, we exploit the machine learning methods in the context of cross-cultural communication, take the advantages of Naive Bayes and support vector machine methods and propose a novel NB-SVM based sentiment analysis algorithm. Extensive experiments show that the proposed approach performs well and can achieve 0:3% error rate of sentiment classification with appropriate parameter settings.
Indexed by:Essay collection
Volume:315-326
Number of Words:10000
Translation or Not:no
Date of Publication:2015-11-30
First Author:Xu Yuemei
Co-author:wangzihou,chenyuji
