One-step learning algorithm selection for classification via convolutional neural networks

• Published in: Information sciences Vol. 721; p. 122610
• Main Authors: Maldonado, Sebastián, Vairetti, Carla, Figueroa, Ignacio
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.12.2025
• Subjects: Algorithm selection; Classifier selection; Machine learning; Meta-learning; Algorithm selection; Classifier selection; Meta-learning; Machine learning
• ISSN: 0020-0255
• DOI: 10.1016/j.ins.2025.122610

As with any task, the process of building machine learning models can benefit from prior experience. Meta-learning for classifier selection leverages knowledge about the characteristics of different datasets and/or the past performance of machine learning techniques to inform better decisions in the current modeling process. Traditional meta-learning approaches first collect metadata that describe this prior experience and then use it as input for an algorithm selection model. In this paper, however, a one-step scheme is proposed in which convolutional neural networks are trained directly on tabular datasets for binary classification. The aim is to learn the underlying structure of the data without the need to explicitly identify meta-features. Experiments with simulated datasets show that the proposed approach achieves near-perfect performance in identifying both linear and nonlinear patterns, outperforming the conventional two-step method based on meta-features. The method is further applied to real-world datasets, providing recommendations on the most suitable classifiers based on the data's inherent structure. •A meta-learning approach for algorithm recommendation is proposed.•Convolutional neural networks are used for automatic classifier recommendation.•A model is constructed with simulated data to learn patterns in tabular datasets.•The approach avoids the construction of meta-features for learning algorithm selection.•The approach outperforms the two-step method based on meta-features.