Proceedings of the Fuzzy System Symposium
41th Fuzzy System Symposium
Session ID : 2G3-3
Conference information

proceeding
Identifying Distinctive Coffee Roasteries Based on Product Descriptions
*Ryozo KitajimaHiroyuki SakaiKeisuke Abe
Author information
CONFERENCE PROCEEDINGS FREE ACCESS

Details
Abstract

In this paper, we aim to identify the most distinctive coffee roasteries in Hokkaido basedon their product descriptions. Coffee has a variety of characteristics―such as sourness, sweetness, andnutty or floral aromas―that are brought out through the roasting process. Given the large number ofroasteries across Japan―estimated at around 5,607―it is likely that some of them reflect the unique stylesor preferences of individual roasters. By analyzing product descriptions written by these roasteries, wefound that the main topics varied. Specifically, three key points were extracted: “flavor,” “coffee origin,”and “brewing method,” and many roasters mentioned flavor. Although the results have some limitations,among those that focus on coffee origin, the roastery that described “farm” and “excellent” was found tobe the most distinctive.

Content from these authors
© 2025 Japan Society for Fuzzy Theory and Intelligent Informatics
Previous article Next article
feedback
Top