Evaluating the Effectiveness of Multiple Machine Learning Models in Date Fruit Recognition

- This study evaluates the efficacy of various machine learning models in classifying date fruits. The objective was to determine the most effective model using comprehensive performance indicators, including AUC, CA, F1, Precision, Recall, and MCC. We employed Random Forest, SVM, Gradient Boosting, AdaBoost, Naive Bayes, and kNN models on a dataset sourced from Kaggle. The results demonstrate that the Gradient Boosting model outperforms others, achieving the highest values in AUC (0.992), CA (0.922), Fl (0.920), Precision (0.927), Recall (0.922), and MCC (0.907). While SVM and Random Forest also showed robust performance, Gradient Boosting established itself as the most reliable for this classification task. However, the study's limitations include the dataset's scope and the specific focus on date fruits. Future research should incorporate additional features and advanced techniques, and validate these models across various agricultural products to enhance generalizability.