“…In light of these research gaps, this study posits the following research questions focusing on the MCG-based e-WOM: This study proposes to extract tweet characteristics from Twitter data posted by the official Twitter accounts of Thai commercial banks, to identify tweet intents from those tweets, to compare the association of intents between RT tweets and FAV tweets, to find the strategies of tweet intents, and to explore the intent strategies together with other characteristics to predict RT or FAV on each tweet, using data-mining techniques (association rules, clustering, and classification). Hamzah and Hidayatullah (2018) tried to cluster Twitter data from the official account of higher education institutions based on hashtags. The results showed that Indonesian higher education institutions mostly used Twitter to post general information, news, agenda, announcement, information for new students, and achievements.…”