Classification of coffee bean categories based upon analysis of fatty acid ingredients

Abstract

The research studies constituents of fatty acids (FA) in coffee beans to identify their categories. Since FA are the fundamental constituents of coffee aroma and flavor, challenges occur to classify the beans' original species in the roasted state. The examined samples in this study cover 74 coffee beans from different origins and are separated into Arabica and Robusta species based on their FA composition. This research develops a discriminant strategy to identify categories of examined coffee beans. This study analyzes an experimental dataset using multiple data structure strategies during the identification process, which are different from traditional approaches that aim to improve coffee bean species classification and recognition rate. Furthermore, the developed coffee bean identification strategies implement various normalization and error analyses during the data reasoning process. This research concludes that FA C18:1, C18:2, and C18:3 own essential characteristics for the coffee beans. Practical applications This research develops an innovative strategy to identify coffee beans in well-known categories and implements various normalization and error analyses during the data processing process. The developed model solves the information loss problem due to data normalization and improves the accuracy of coffee been classification. Consequently, the study concludes that fatty acids C18:1, C18:2, and C18:3 characterize essential features for coffee beans, and the research results balance fundamental chemistry and engineering principles and serve the general food processing and preservation technology industries.