TY - JOUR
T1 - Applying sentiment analysis to automatically classify consumer comments concerning marketing 4Cs aspects
AU - Lin, Hao Chiang Koong
AU - Wang, Tao Hua
AU - Lin, Guo Chung
AU - Cheng, Shu Chen
AU - Chen, Hong Ren
AU - Huang, Yueh Min
N1 - Publisher Copyright:
© 2020
PY - 2020/12
Y1 - 2020/12
N2 - With the rapid development of science and technology, consumers are used to searching online for evaluations before purchasing products. Manufacturers can also utilize such information like users’ usage habits, browsed websites, comments, messages, etc. to formulate marketing strategies suitable for their products. Several researches developed opinion mining on predicting the polarity of consumers’ comments, but few of them were from marketing point of view. In this regards, this study looks to establish an automated way to collect and analyze consumers’ comments in social networks, automatically classify them into marketing 4Cs and non-marketing categories from a large number of consumer comments, and divide the category of marketing 4Cs articles into four types of attribute dimensions to analyze emotional polarity. Based on the marketing theory of 4Cs and LDA topic analysis, this study extracted the characteristic keywords from the collected consumer reviews for corpus classification and sentiment polarity analysis. This study further establishes a feature keyword library for specific fields, hoping to improve the accuracy of sentiment analysis through these keywords, simplify the process of consumers’ searches for product evaluations, and facilitate consumers to search for helpful target information.
AB - With the rapid development of science and technology, consumers are used to searching online for evaluations before purchasing products. Manufacturers can also utilize such information like users’ usage habits, browsed websites, comments, messages, etc. to formulate marketing strategies suitable for their products. Several researches developed opinion mining on predicting the polarity of consumers’ comments, but few of them were from marketing point of view. In this regards, this study looks to establish an automated way to collect and analyze consumers’ comments in social networks, automatically classify them into marketing 4Cs and non-marketing categories from a large number of consumer comments, and divide the category of marketing 4Cs articles into four types of attribute dimensions to analyze emotional polarity. Based on the marketing theory of 4Cs and LDA topic analysis, this study extracted the characteristic keywords from the collected consumer reviews for corpus classification and sentiment polarity analysis. This study further establishes a feature keyword library for specific fields, hoping to improve the accuracy of sentiment analysis through these keywords, simplify the process of consumers’ searches for product evaluations, and facilitate consumers to search for helpful target information.
UR - http://www.scopus.com/inward/record.url?scp=85092400509&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092400509&partnerID=8YFLogxK
U2 - 10.1016/j.asoc.2020.106755
DO - 10.1016/j.asoc.2020.106755
M3 - Article
AN - SCOPUS:85092400509
VL - 97
JO - Applied Soft Computing
JF - Applied Soft Computing
SN - 1568-4946
M1 - 106755
ER -