Sentiment Analysis of Brand Personality Positioning Through Text Mining

Sentiment Analysis of Brand Personality Positioning Through Text Mining

Ruei-Shan Lu (Takming University of Science and Technology, Taipei, Taiwan), Hsiu-Yuan Tsao (National Chung Hsing University, Taiwan), Hao-Chaing Koong Lin (National University of Tainan, Tainan, Taiwan), Yu-Chun Ma (National University of Tainan, Tainan, Taiwan) and Cheng-Tung Chuang (Takming University of Science and Technology, Taipei, Taiwan)
Copyright: © 2019 |Pages: 11
DOI: 10.4018/JITR.2019070106

Abstract

This article uses text mining and a Chinese word segmentation program developed by the Chinese Knowledge and Information Processing Group in Taiwan's Academia Sinica to analyze Facebook posts from 14 e-commerce companies. In addition, a list of keywords representing brand personalities is analyzed to reveal key factors affecting which social media posts attract consumers' attention. This research uses statistical analysis with a nonmanual questionnaire that is efficient and based on computer science to provide a reference for businesses operating Facebook fan pages and internet marketing.
Article Preview
Top

Introduction

As internet and social media develops, the usage of smart mobile devices increases, the users of internet increases while the user’s age bracket decreases. The popularity of Social Networking Site and Instant Messaging lead many business companies to interact and promote its products to consumers via Facebook, Twitter and other social networking sites, sometimes on instant messaging apps such as Line, QQ and WeChat. In the “2013 Taiwan’s Internet User Shopping Behavior Analysis” by Market Intelligence & Consulting Institute (MIC) points out that in the category “Which way to buy needed products,” “Browse social network sites” takes up 30.1%, which is the second highest in all option. The fan pages in social networking sites always takes up 18.1%. From this statistic analysis, it’s clear that the information in social networking sites cannot just only raise Taiwan’s internet user’s shopping desire, it is also a main source when it comes to comparing products.

According to “Facebook Taiwan Consumer Online Behavior Questionnaire,” a research that Facebook commissioned Kantar TNS to conduct points out the fact that Facebook is the most used social network site by Taiwanese people. Consumers, especially students and young professionals spends more time on Facebook than any other websites, TV and printed media. Facebook has a huge influence on Taiwanese consumer, 50% of Facebook Taiwan consumer searched, bought or sold products on Facebook, 22% of Taiwanese that used Facebook will buy the product after getting recommendations from friends. The sales conversion rate is double the 12% of self-search. From this report, it’s not hard to realize that Facebook is not just a simple social networking site, but a battleground of business marketing. How to promote and popularize product and corporate identity through Facebook become a something that all business corporate must think about.

This research hopes to use text mining as the main technique, with the data from “Taiwan’s Top 10 Online Shopping Sites With The Most Product” provided by EZprice in July 2014, the Facebook fan pages of 10 shopping sites “PChome Online Shopping,” “Momo Shop,” “Yahoo Shopping Center,” “UDN Buy,” “Go Happy Shopping Site,” “PayEasy Women Shop,” “ETMall,” “U-Mall,” “ 7net Online Shopping,” “ASAP Flash Online Shopping” as the research objective, using fan page posts from the day the fan page was created until 2016 as the field of research. Through the list of vocabularies that represent brand personality positioning organized from papers and documents, also by calculating the brand personality positioning emotion score, the result can be used in corporate’s Facebook fan pages as reference when it comes to marketing and brand market position.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 13: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 12: 4 Issues (2019)
Volume 11: 4 Issues (2018)
Volume 10: 4 Issues (2017)
Volume 9: 4 Issues (2016)
Volume 8: 4 Issues (2015)
Volume 7: 4 Issues (2014)
Volume 6: 4 Issues (2013)
Volume 5: 4 Issues (2012)
Volume 4: 4 Issues (2011)
Volume 3: 4 Issues (2010)
Volume 2: 4 Issues (2009)
Volume 1: 4 Issues (2008)
View Complete Journal Contents Listing