Optimizing PROAFTN classifier with ant colony algorithm: enhanced diabetes detection benchmarking

The increasing global prevalence of diabetes highlights the need for accurate diagnostic tools to improve early detection and effective treatment planning. Traditional classification models often struggle to achieve optimal performance due to limitations in parameter tuning and adaptability to complex datasets. To address these limitations, this article introduces PROAnt, an innovative learning approach designed to enhance the robustness and efficiency of the PROAFTN multicriteria classification method. PROAnt leverages the computational power of ant colony optimization (ACO) to dynamically fine-tune and optimize the key parameters, such as intervals and weights, at the core of the PROAFTN classification process. This learning methodology is crucial because PROAFTN depends on these parameters for classification. ACO, inspired by the foraging behavior of ants, achieves rapid and accurate convergence with minimal parameters, outperforming many traditional optimization techniques. This study demonstrates how ACO can inductively infer PROAFTN’s parameters from data, leading to high accuracy and precision. Evaluations were conducted on a publicly available diabetes dataset containing 100,000 samples, sourced from Kaggle’s 2023 machine learning competition. This dataset includes demographic and clinical features, providing a robust basis for model benchmarking. The model achieved a classification accuracy of 98.42%, along with a weighted precision of 83.73%, a weighted recall of 0.789, and a weighted kappa of 0.789 ± 0.008, outperforming other baseline classifiers such as a feedforward deep neural network (three hidden layers: 128, 64, 32), decision trees, k-Nearest Neighbors (k-NN) and logistic regression. These findings underscore the transformative potential of integrating ACO with PROAFTN, not only in advancing diabetes detection but also in the broader application of artificial intelligence.