Predicting Apple Quality Using Machine Learning Techniques

In order to optimize the agricultural supply chain and ensure customer satisfaction, evaluating apple quality is crucial. The use of machine learning (ML) algorithms offers a viable solution to the subjective and time-consuming nature of traditional quality assessment methods. To classify apples as "good" or "bad" quality, the dataset used in this study includes thorough evaluations of relevant characteristics. This study investigates the use of multiple machine learning models to assess apple quality based on characteristics such as size, weight, sweetness, acidity, ripeness, juiciness, and crunchiness. To build predictive models, machine learning techniques such as Naïve Bayes, decision trees, K-nearest neighbors (KNN), and AutoMLP were employed. Among these models, AutoMLP achieved the highest accuracy rate of 88.46%.These results provide a foundation for future advancements in fruit quality assessment and contribute to the progress of agricultural technologies.