Assessing Destination Brand Associations on Twitter: The case of Istanbul

Abstract

The development of data mining has paved the way for studies that identify brand associations from user-generated content (UGC). However, the number of studies investigating destination associations with social media is limited. The aim of this study is to explore destination associations with UGC on Twitter and to show how data mining and sentiment analysis methods can be applied to destinations to elicit brand associations. In this study, 33,339 English-language tweets containing the word #Istanbul were collected over one year and analyzed using text mining (association rule analysis) and sentiment analysis. As a result of the study, a brand concept map (BCM) of what Twitter users associate with Istanbul was created and compared to other studies that measure associations using conventional methods. The main results show that users have positive associations with tourism in Istanbul. Unique and interesting associations (such as "cats") were observed compared to other previous studies that measured associations to destinations. Based on the study results, a method was proposed for measuring the image of a place brand by observing electronic word of mouth in social media.