A Sonification of Cross-Cultural Differences in Happiness-Related Tweets

We report an approach to communicating cross-cultural differences in sentiment data through sonification. Sonification is a powerful technique for the translation of patterns into sound that is understandable, accessible, and musically pleasant. A machine-learning classifier was trained on sentiment information of two samples of Tweets with the keyword of “happiness,” from Singapore and New York. Positively valenced words, which we argue relate to the concept of “happiness,” showed stronger influences on the classifier than negative words. For mapping, Tweet frequency differences of the semantic variable “anticipation” affected tempo, “positive” affected pitch, “joy” affected loudness, and “trust” affected rhythmic regularity. We evaluated a sonification of the original two cities’ data, together with a control condition generated from random mappings, in a listening experiment. Results suggest that the original was rated as significantly more pleasant. No effects were found on ratings of interest or confidence in hearing a difference. We claim that our statistically informed approach to sonification enabled the making of better aesthetic decisions in the design of a pleasant listening experience. Further studies might employ extensive sentiment analysis of social media data and sonification with more input parameters and facilitate the communication of statistical results and a nuanced understanding of cross-cultural differences.