BEER AND MACHINE LEARNING: RECOMMENDING BEER STYLES

Abstract

This work aimed to solve the problem faced by some people who struggle to appreciate special beers due to a lack of guidance on what to choose. To achieve this goal, studies were conducted involving Euclidean distance and the Term Frequency-Inverse Document Frequency (TF-IDF) method to find similarities between beer styles, based on a specific style. Additionally, the vast Brazilian beer landscape was explored, and the BJCP guide, which served as the study’s foundation, was presented as a directional resource for better understanding beer styles. Through three tests, the proposed methods were validated, resulting in the development of a tool capable of suggesting beer styles based on a reference style.