Applying sentiment analysis to automatically classify consumer comments concerning marketing 4Cs aspects

Hao-Chiang Koong Lin,Tao-Hua Wang,Guo-Chung Lin,Shu-Chen Cheng,Hong-Ren Chen,Yueh-Min Huang
DOI: https://doi.org/10.1016/j.asoc.2020.106755
IF: 8.7
2020-12-01
Applied Soft Computing
Abstract:<p>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.</p>
computer science, artificial intelligence, interdisciplinary applications
What problem does this paper attempt to address?
The main problem this paper attempts to address is the automated collection and analysis of consumer reviews on social media, and the automatic classification of these reviews into the marketing 4Cs (Consumer, Cost, Convenience, and Communication) and non-marketing categories. Specifically, the research objectives include: 1. **Automated Collection and Analysis of Consumer Reviews**: Extract useful information from a large number of consumer reviews through automated means to help manufacturers better understand consumer needs and opinions. 2. **Sentiment Polarity Analysis**: Perform sentiment analysis on the collected reviews based on the 4Cs attributes in marketing theory to understand consumers' positive or negative evaluations of different aspects of the product. 3. **Improving Search Efficiency**: Simplify the process for consumers to find product reviews, allowing them to quickly locate the target information they need. 4. **Establishment of a Feature Keyword Library**: Construct a feature keyword library for specific fields, aiming to improve the accuracy of sentiment analysis through these keywords. 5. **Topic Model Analysis**: Use the LDA topic model to conduct preliminary topic analysis of the reviews, reducing the manpower and time costs in the subsequent establishment of the feature word library. Overall, this paper aims to utilize sentiment analysis technology to enhance the automation of consumer review processing, thereby supporting enterprises in formulating effective marketing strategies and providing consumers with more convenient ways to access information.