Mining and exploring electronic word-of-mouth from Twitter: case of the Java Jazz Festival

Abstract

Purpose This paper aims to explore textual patterns in ten years of electronic word-of-mouth communications amongst social media (SM) users of the Java Jazz Festival. Design/methodology/approach This study uses a data-scraping technique to gather user-generated content from Twitter. Word-cloud and word-frequency analysis, along with descriptive coding and pattern matching, are used to categorise the initial findings. Trends and differences in terms of the number of tweets over a ten-year period were examined using analysis of variance and seasonality analysis. Findings From more than 1.3 million Twitter tweets between 2008 and 2018, this study identified six initial themes. Quantitative analysis revealed that the number of tweets differed significantly in the four quarters of the ten-year period. Research limitations/implications The results of this study contrast with the claim that digital media communication generally occurs before a festival begins and are least during the festival. Nevertheless, this study supports the notion that SM interaction results in positive consequences, drives conversations amongst users and increases engagement. Practical implications This study offers five practical implications for music festival organisers and related entities. Originality/value To the best of the author’s knowledge, this study is the first to provide a systematic and practical data mining and interpretation approach from Twitter within a ten-year period in the Asia Pacific context, through the case of the Java Jazz Festival.