Classification of Cocoa Beans Based on the Grade Level of Fermentation using KNN Algorithm

Abstract

Objectives: This study aims to develop a cacao bean fermentation prototype, specifically this study covers: (1) Development of cacao bean fermentation device using Arduino microcontroller, (2) Creating a model that will be used in evaluating the fermented cocoa bean using KNN Algorithm, and (3) Evaluating the developed system using the ISO 9126 software quality model in terms of functionality, reliability, usability, efficiency, maintainability, and portability. Methodology: The researcher collected, presented, and analyzed the necessary data and information to address the research objectives. Reasons and justifications for the Developmental Model, Gantt chart, Data Gathering Procedure and Technique, Context Diagram, Data and Process Model, and Schematic Diagram are given. Findings: This study aimed to address these issues by automating the cacao bean fermentation process. A novel device was created and assessed using ISO 9126 and the K-Nearest Neighbor algorithm. The results demonstrated excellent performance in functionality (4.77), reliability (4.56), usability (4.58), efficiency (4.6), maintainability (4.6), and portability (4.56). The device effectively mixed cacao beans during fermentation and achieved an 85.3% accuracy rate in image classification. Novelty: In comparison to existing methods, the device surpassed expectations in terms of fermentation control and software reliability. Overall, the developed Cacao Bean Fermentation Device significantly enhances functionality and control, ensuring consistent quality in chocolate production. Keywords: Arduino Microcontroller, KNN algorithm, Image Analysis, Cacao Bean Fermentation