Authors: Joyce Lee, Shu-Fen Tseng and Shih-Yun Chen
Social media full of user opinions, experiences, wishes and complaints have been considered as being a gold mine for extracting business knowledge. However, the constant accumulation of user opinions and comments means that capturing insightful business knowledge is a challenging task. This is especially the case because business know-how in this context is seldom explicitly obvious, but rather, implicit in its nature as well as being unstructured and hence, difficult to interpret. Particularly, the quirky expressions created by the users make the messages hard to read for people who want to know “what is happening” and “what is going on” on the forums in relation to their business operations. We argue that in order to understand better what the web users say and mean, it is critically important that researchers and practitioners strive systematically to interpret both the semantic (meanings of words) and pragmatic (context of words) meanings of the content. Accordingly, this research involves unpacking the textual meaning of interactions in social media. For this research, we intend to tackle the challenges faced by individuals and organizations in trying to deal with “big data”, with the focus on the textual meaning and context of social media. This study focuses on the automobile industry. One statistical report has shown that there is an increasing need for car buyers and suppliers to engage with social media so as to keep abreast of consumer trends; however, many companies and manufacturers struggle with how to convert such big online data into useful business knowledge. In order to address this, we conduct network text analysis (Popping, 2000) based on the data collected from the most popular online discussion forum in Taiwan, Mobile01 (www.mobile01.com), from which 10 of the longest car-related discussions dated from February 2007 to September 2014 (approximately 14,000 posts) are collected for data analysis.The preliminary findings reveal that (1) a large amount of web slang containing irony and/or sarcasm has been created by the users and (2) these often connote sentimental meanings that are commonly adopted in the forum and hence, become generic terms. (3) This results in the creation of a collective opinion climate towards a product and/or service. Moreover, the outcomes demonstrate that whilst valuable business insights are concealed in such big reserves of data, benefits can accrue from small data derived from a text and network approach. This allows for exploration of the meanings contained in social media in a longitudinal manner, thus contributing a new perspective to studies in the field.
For the future study, we intend to establish a framework for converting unstructured expressions into structured formats and hence, contribute to what other prior scholars have described as moving from being “data-rich” to “insight-rich” and responding effectively to the best of these insights.
