Performance Evaluation of Classification Multi Algorithms on Small Dataset: A Comparative-Based Analysis

This research aims to evaluate and compare the performance of various classification algorithms under conditions of small dataset quantity. Six algorithms were tested, namely NaÏve Bayes, Support Vector Machine (SVM), Decision Tree, Random Forest, AdaBoost, and k-NN against three small-scale public datasets from different domains. The experiments were conducted using cross-validation techniques (k-fold=5), and evaluations were performed based on the confusion matrix for performance accuracy, precision, recall, F1-score, AUC, and model computation time. The results of the study show that the Naive Bayes algorithm excels in all feature number datasets. The SVM and Random Forest algorithms excel in all feature text, while Random Forest and Naive Bayes excel in mix feature. Each user should choose an algorithm that is tailored to the data used. Additionally, the use of Cross-Validation has proven to provide a more reliable performance estimation. This finding provides practical recommendations for researchers and practitioners in selecting effective classification algorithms for small-scale datasets, as well as highlighting the importance of validation techniques and data processing in improving model generalization under data limitations.