One of the most consumed beverages in the world is tea, whether due to its health benefits or its unique flavors. With the advancement of artificial intelligence, it shows promise for use in sensory analysis. Machine learning algorithms can be used to identify correlations between the sensory characteristics of teas and consumer preferences, enabling the market to meet demands more accurately and effectively. The objective of this study was to apply machine learning techniques to build a recommendation system for commercial teas based on similarities among consumers. The commercial teas used were: chamomile, fruits and flowers, mixed (herbs, flowers, and fruits), and red flowers. A total of three hundred fifty evaluators participated, conducting an acceptance test with a five-point hedonic scale and a questionnaire including gender and age group. The Python programming language was used, along with the KNN (k-nearest neighbors) model and the similarity function to create the User Similarity Matrix. The results indicated that the most preferred tea was chamomile, followed by mixed, fruits and flowers, and lastly red flowers. The matrix showed that the highest similarity values, from zero point ninety-six to zero point ninety-five, for preference towards chamomile tea were found among females aged eighteen to twenty-five years and males aged twenty-six to thirty-nine years. The recommendation prediction for each user indicated that for the other teas, the preference scores would range between four point zero and two point zero. Therefore, it was concluded that the combination of these two fields of knowledge can make the industry more dynamic and adaptable to consumer preferences, as well as contribute to more targeted marketing in product recommendations.