Towards Optimizing Espresso Extraction: Which Perceived Espresso Features Correlate with Overall Subjective Satisfaction?

Coffee is one of the most frequently consumed drink beverage in the world. It is an essential part of our daily life. However, not much study has been done into optimizing coffee extraction with regards to user satisfaction. Therefore, in this study we present a new approach that integrates statistics and data science techniques to understand the interaction between various parameters involved in coffee extraction and consumer preferences in the context of coffee extraction, specifically the espresso type coffee. In the method we collect objective information using a DE1PRO espresso machine and subjective information from consumers through an online survey. The final goal of this work is to develop an espresso extraction model that would both fit consumer preferences while allowing for a higher sustainability of coffee industry. The initial statistical analysis presented in this paper shows that features like aroma, bitterness, concentration and richness are affecting the overall satisfaction of consumers to a different degree. We also found out that perceived richness of taste is the one parameter that was the most correlated with overall satisfaction, even more than aroma. On the other hand, the subjectively perceived strength of espresso beverage was the least correlated feature, although the correlation was not statistically significant. Future experiments will include sensory information in the analysis and implement causal inference for better insight into relationship between subjectively perceived satisfaction and objective information.