Journal overview and metrics

Aims

MLCS is an international, peer-reviewed, open-access journal dedicated to research in which machine learning and a substantive application domain advance together. The journal aims to close the distance between methodological innovation and the settings where such methods must eventually work, including clinics, farms, schools, public agencies, and industry, by bringing computer scientists, domain specialists, and practitioners into a shared literature. MLCS prioritises submissions that combine methodological rigour with demonstrable consequences for practice, policy, or scientific understanding within the domain they address.

Convergence, as this journal uses the term, means the genuine integration of computational methods with domain knowledge: the problem is formulated together with the domain, the data reflect its real operating conditions, the evaluation reflects the decisions at stake, and the findings remain interpretable to the people who will act on them.

Scope

We invite original research articles, review articles, and short communications that address demanding problems in science, health, and society through advanced computational methods. The journal's scope rests on three pillars.

1. Machine Learning Methodology: the development and rigorous evaluation of learning methods, particularly under the constraints real applications impose.

• Learning under real-world constraints: imbalanced, small-sample, noisy, and incomplete data; transfer, few-shot, and semi-supervised learning.
• Architectures and representation: deep learning architectures, representation learning, sequence and time-series modelling, multimodal and graph-based learning.
• Feature engineering and model selection: feature selection, dimensionality reduction, hyperparameter optimisation, ensemble design.
• Evaluation methodology: validation design, uncertainty quantification, calibration, statistical comparison of models, and reproducible benchmarking.
• Optimisation and bio-inspired computation: evolutionary algorithms, swarm intelligence, and metaheuristics for learning and resource allocation.

2. Convergence Domains: studies in which a learning method is developed, adapted, or validated against the conditions of a specific field.

• Health and clinical sciences: clinical decision support, diagnostic and prognostic modelling, medical imaging and biosignal analysis, nursing and public health informatics, epidemiological forecasting.
• Food systems and agriculture: yield and supply forecasting, quality and grading classification, precision agriculture, food safety analytics.
• Development, society, and public policy: poverty measurement and village development indices, spatial and socio-economic modelling, education analytics, and evidence for public decision-making.
• Environment, energy, and infrastructure: remote sensing, disaster risk assessment, demand forecasting, and condition monitoring of engineered systems.

3. Trustworthy and Deployable Intelligence: the conditions under which an intelligent system can responsibly be relied upon.

• Explainability: interpretable models and explanation methods evaluated with the domain experts who must use them.
• Fairness and social consequence: bias measurement and mitigation, distributional impact of automated decisions.
• Robustness: distribution shift, adversarial and failure analysis, post-deployment monitoring.
• Privacy and governance: privacy-preserving and federated learning, data stewardship, and the ethics of automated decision-making.
• Deployment: edge and IoT computing, learning under limited computational and data infrastructure, and human–machine collaboration.

Outside the scope. Studies applying an established method to a new dataset without methodological contribution or substantive domain insight; purely theoretical results with no demonstrated application; and manuscripts whose domain claims are not supported by expertise within the author team.

Alignment with the UN Sustainable Development Goals

MLCS is committed to research that serves development as well as scholarship. By publishing work on intelligent diagnostics, digital health systems, and epidemiological analytics, the journal supports SDG 3 (Good Health and Well-being). Through research on yield forecasting, precision agriculture, and food safety, it contributes to SDG 2 (Zero Hunger). Its emphasis on poverty measurement, village development indices, and data-driven public decision-making advances SDG 1 (No Poverty). And by developing robust, explainable, and deployable computational methods, including for settings with limited infrastructure, MLCS contributes to SDG 9 (Industry, Innovation and Infrastructure).