Text Mining Analysis of Q-Grader Sensory Descriptors in Specialty Coffee Under Accelerated Storage Conditions

Area of Science:

• Food Science
• Sensory Analysis
• Computational Linguistics

Background:

• Specialty coffee quality assessment relies on sensory evaluation, but expert descriptive narratives are underutilized.
• Understanding coffee quality evolution during storage is crucial for the industry.

Purpose of the Study:

• To develop an integrated analytical framework combining sensory evaluation and NLP.
• To characterize specialty coffee quality changes during accelerated storage under various packaging systems.

Main Methods:

• Accelerated storage of coffee at 40, 50, and 60 °C in eight packaging configurations.
• Sensory evaluation using the Specialty Coffee Association protocol.
• Textual sensory description analysis using NLP techniques (descriptor frequency, TF-IDF, topic modeling).

Main Results:

• Packaging-product configurations and temperature significantly impacted sensory stability.
• Vacuum packaging and multilayer bags maintained desirable attributes and higher cup scores.
• Elevated temperatures and grinding accelerated deterioration, replacing positive descriptors with negative ones.

Conclusions:

• Integrating NLP with sensory evaluation transforms qualitative narratives into quantitative data.
• This framework provides objective characterization of sensory changes during coffee storage.
• NLP enhances the comprehensive assessment of specialty coffee quality evolution.