Skip to content
Open access

RF-SPEA2 Hybrid Rule Inference and Optimization Architecture

Sep 2026 · Applied Sciences · Vol 16, pp. 9593 · 0 citations · 7 references

TL;DR

Competitive predictive performance against conventional machine learning models while outperforming several traditional rule-based approaches is achieved and a promising framework for combining competitive predictive performance with an explicit rule-based representation across datasets with different dimensionalities and class structures is provided.

Abstract

The high predictive performance of machine learning models in complex classification tasks often comes with limited transparency, particularly for black-box models whose decision processes cannot be directly inspected. In this study, a two-stage Random Forest–SPEA2 hybrid rule inference and optimization architecture is proposed to achieve high predictive performance while retaining an explicit rule-based decision structure. In the first stage, root-to-leaf decision paths extracted from Random Forest trees are converted into interval-based IF–THEN rules, and a class-balanced candidate rule pool is constructed. In the second stage, the Strength Pareto Evolutionary Algorithm 2 (SPEA2) simultaneously maximizes rule-level accuracy and weighted F1-score using Pareto-based multi-objective optimization. The resulting rules preserve an explicit IF–THEN representation whose feature conditions and decision boundaries can be directly inspected. The proposed framework was evaluated on three benchmark datasets representing different classification characteristics: Breast Cancer Wisconsin for binary classification, Dry Bean for multi-class classification, and ISOLET for high-dimensional multi-class classification. Under stratified 5-fold cross-validation, RF-SPEA2 achieved mean accuracies of 96.77%, 98.08%, and 95.54% on Breast Cancer Wisconsin, Dry Bean, and ISOLET, respectively, with relatively low variability across folds. Comparative experiments showed that the proposed method achieved competitive predictive performance against conventional machine learning models while outperforming several traditional rule-based approaches. Representative IF–THEN rules obtained for the three datasets demonstrate that the resulting decision structure remains directly inspectable, although the practical interpretability of individual rules depends on the dimensionality and semantic meaning of the input features. These findings indicate that RF-SPEA2 provides a promising framework for combining competitive predictive performance with an explicit rule-based representation across datasets with different dimensionalities and class structures.

Read PDF

Similar papers

Open access Oct 2026

DP-FDT: The Drug Pair-Based Fuzzy Decision Tree for Accurate Drug Combination Prediction

This paper presents a novel algorithm, the Drug Pair-based Fuzzy Decision Tree (DP-FDT), designed to improve classification performance on imbalanced drug combination datasets. Unlike traditional decision trees, the DP-FDT integrates drug combination feature pairs into its feature selection and node construction proces...

Zhen-Hua Cai, Jing Chen, Kun-Hong Liu · 0 citations
Open access Sep 2026

A hybrid NSGA-II and random forest framework for multi-objective credit risk assessment

Credit risk assessment is an important way to measure the likelihood of loan default. It helps institutions stay solvent and manage their capital effectively, which supports strong risk management in the financial sector. This study presents a new hybrid predictive framework that combines the Non-dominated Sorting Gene...

Alaa S. Alaerjan, Ayman Mohamed Mostafa, Mayada Tarek et al. · 0 citations
Open access Oct 2026

Generalized Gradient of Positive Adaptive-Compensation-Based for Fuzzy Hierarchical Classification

Hierarchical Takagi–Sugeno–Kang (TSK) fuzzy classifiers combine nonlinear modeling with rule-based decision structures for complex classification tasks. However, prediction errors may propagate through successive layers, and the contributions of different layers require careful consideration when combining their output...

Yue Xin, Wen-Liang Li, Ta Zhou · 0 citations
Open access 2026

Explainable heterogeneous ensemble learning with hybrid feature selection for heart disease prediction

The paper introduces an ensemble learning framework for predicting heart disease with a combination of the pre-processing, hybrid feature selection, class-imbalance management, hyper-parameters optimization, ensemble learning, and Explainable Artificial Intelligence (XAI). For feature selection, the presented framework...

Saba Fatima, M. Arasakumar, P. Anisha · 0 citations
Open access Oct 2026

Optimizing Classification Algorithms With Smote And Metaheuristic Optimization For Imbalanced Breast Cancer Data

Class imbalance remains a critical challenge in medical classification, particularly in breast cancer diagnosis, where misclassification of the minority class can have severe clinical consequences. This study investigates the impact of combining the Synthetic Minority Over-sampling Technique (SMOTE) and metaheuristic o...

Yusup Maolani, Samidi Samidi · 0 citations
Conference Sep 2026

Intelligent investment project evaluation system-based on ensemble learning and Bayesian optimization: multicriteria decision algorithm for investment decision efficiency

The pattern-based risk classification approach for the level of investment projects involves the following two major challenges: low discriminant efficiency and poor separability of features in high dimensional spaces. Traditional methods demonstrate significant computation barriers in dealing with large and heterogene...

Ting-Ting Zhao, Xue-Chun Wang · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.