scikit-learn

The algorithm selection and parameterization (ASP) domain lacks systematic surveys and empirical evaluations. Method: We propose the first standardized, meta-learning–driven ASP framework, built upon the largest ASP benchmark knowledge base to date—comprising 400 datasets and 4 million pre-trained models—and conduct large-scale comparative experiments across eight mainstream classifiers under diverse scenarios. Our evaluation integrates empirical performance modeling (EPM), feature engineering, and statistical significance testing to quantify accuracy, generalizability, and computational efficiency. Contribution/Results: This work delivers the first critical survey balancing methodological rigor with empirical breadth; reveals performance boundaries and applicability conditions of state-of-the-art ASP methods; and establishes a reproducible benchmark and practical selection guide for AutoML research and deployment.

Biomedical machine learning modeling often suffers from high computational resource consumption, poor code reusability, and insufficient reproducibility and traceability. To address these challenges, we propose an end-to-end, interpretable, lightweight ML workflow that innovatively integrates scikit-learn, MLflow, and SHAP—enabling automated experiment tracking, model training, post-hoc interpretability analysis, and modular extensibility. Designed as a template-based framework, it supports seamless cross-project transfer, significantly improving modeling efficiency, result reproducibility, and team collaboration. The workflow is open-sourced and has been adopted by multiple bioinformatics teams for disease prediction and multi-omics analysis tasks. It establishes the first standardized, production-ready ML engineering practice tailored to biomedical research, bridging a critical gap between methodological innovation and scalable, transparent, and maintainable ML deployment in the domain.

To address the low training efficiency and excessive computational cost of traditional machine learning algorithms on large-scale datasets, this paper proposes a high-efficiency training framework grounded in mathematical reformulation and numerical optimization. By rigorously rederiving and restructuring the core optimization procedures of classical algorithms—including logistic regression, support vector machines (SVM), and linear regression—the framework eliminates redundant computations and integrates sparse acceleration with low-rank approximations, all while preserving theoretical accuracy. Experiments demonstrate significant speedups: up to 4× faster than scikit-learn on multivariate linear regression, logistic regression, and SVM tasks; and as high as 800× acceleration for SVM on large-scale datasets, with zero loss in predictive accuracy. The framework unifies support for regression, classification, clustering, and dimensionality reduction, offering both generality and scalability. It establishes a new paradigm for large-scale machine learning that simultaneously ensures high precision and high computational efficiency.

This paper identifies a systemic issue in machine learning: preprocessing hyperparameters—such as missing-value imputation strategies—are frequently overlooked yet substantially bias model evaluation. Current practice often involves informal, post-hoc tuning of preprocessing steps, leading to optimistic performance estimates and irreproducible results. To address this, the authors formally distinguish and empirically analyze the coupling effects between algorithmic and preprocessing hyperparameters. Using a modular supervised learning workflow model, controlled variable experiments, replication of canonical case studies, and bias diagnostics, they quantify the resulting optimistic bias. Key contributions include: (1) establishing preprocessing hyperparameters as equally critical as algorithmic ones; (2) proposing formal modeling principles to eliminate informal preprocessing tuning; and (3) delivering actionable reporting guidelines for ML practitioners, thereby significantly enhancing model credibility and reproducibility.

Machine learning model selection lacks formalized methodologies, making it difficult to systematically characterize contextual factors—such as data characteristics and prediction tasks—and their interactions, resulting in opaque, non-adaptive decisions. This paper introduces, for the first time, software product line (SPL) principles into ML model selection, proposing a variability-aware algorithm selection framework. It constructs a configurable feature model that explicitly captures commonalities and variabilities among contextual factors—including dataset size, feature dimensionality, and task type—as well as their logical dependencies. By integrating scikit-learn’s heuristic rules with an instantiation framework, the approach enables interpretable, adaptive, and transparent model recommendations. An empirical case study demonstrates that the method significantly outperforms existing strategies in accuracy, interpretability, and contextual adaptability.

Scientific software often suffers from poor reproducibility and low sharing rates, primarily because researchers lack formal software engineering training—leading to version chaos, uncontrolled code quality, and cumbersome release processes. To address this, scikit-package introduces a progressive software engineering roadmap tailored for domain scientists. It provides standardized packaging tutorials, automated workflow templates (covering build systems, CI/CD pipelines, documentation generation, and package management), and community-maintained pedagogical resources. Its key innovation lies in adapting professional software engineering practices to the cognitive load of non-specialist programmers, enabling systematic progression from script-based functions to production-ready, open-source package releases. Empirical evaluation demonstrates that scikit-package significantly improves the reproducibility and maintainability of scientific code, enhances community sharing efficiency, and lowers barriers to standardized scientific software publication.

This study addresses the limitations of existing approaches for automatically extracting machine learning (ML) pipeline structures, which often rely on manual annotations or suffer from insufficient generalization to keep pace with the rapid evolution of the ML ecosystem. The work presents the first systematic evaluation of small language models (SLMs) for reverse-engineering ML pipelines and proposes an SLM-based method for their automatic identification and reconstruction. Through comprehensive comparative experiments across multiple SLMs and rigorous statistical validation using Cochran’s Q, McNemar, and Pearson’s chi-squared tests, the authors demonstrate that the best-performing SLM significantly outperforms current methods and exhibits robustness across diverse classification schemes. This approach uncovers finer-grained patterns in data science practices and overcomes longstanding bottlenecks in scalability and domain adaptability inherent in traditional techniques.

This study addresses the dual limitations of weak interpretability in machine learning (ML) models and insufficient flexibility in classical statistical models. Methodologically, we propose a modular hybrid architecture that tightly integrates ML components—such as tree-based models and neural networks—with statistical frameworks—including generalized linear models and Bayesian inference—thereby preserving statistical interpretability while enhancing nonlinear modeling capacity. Our key contributions are: (1) a unified modeling framework that reconciles theoretical tractability with data-driven adaptability; and (2) improved robustness via structured regularization and uncertainty propagation. Extensive experiments on heterogeneous, multi-source datasets demonstrate that our approach achieves an average 8.3% improvement in predictive accuracy over both pure-ML and pure-statistical baselines, while significantly enhancing decision trustworthiness and model debugging efficiency.

This work addresses the high annotation costs and inefficiency of existing AI tools in tabular software engineering optimization tasks by introducing EZR.py, a lightweight Python toolkit. By re-engineering classical algorithms, EZR.py uncovers profound structural similarities among diverse learning and optimization methods—including Naïve Bayes, CART, k-means, simulated annealing, and active learning—and unifies their implementation in just a few hundred lines of code. Evaluated on over 120 tasks from the MOOT benchmark suite, EZR.py matches or exceeds the performance of established tools such as SHAP, LIME, SMAC3, and FASTREAD, while achieving up to 500× faster runtime, reducing label requirements by several orders of magnitude, and relying on fewer than ten input variables—thereby substantially lowering model complexity and computational overhead.

This work proposes the first end-to-end automated artificial intelligence research framework capable of fully automating the development pipeline from algorithmic idea generation to executable machine learning classifiers. The approach integrates structured meta-prompt engineering with large language model–based code generation, augmented by an automated evaluation and iterative optimization mechanism. Experimental results on twenty standard datasets from the Infinity-Bench benchmark demonstrate that multiple novel classifiers autonomously generated by the framework significantly outperform baseline methods implemented in scikit-learn. This study thus achieves, for the first time, complete automation of the entire workflow—from initial algorithmic conception to deployable, runnable code—marking a significant step toward self-driving AI research systems.

This work addresses the limitations of existing machine learning prototyping tools, which often lack effective support for collaboration and cross-project knowledge reuse, leading to tool fragmentation and insufficient stakeholder engagement. To overcome these challenges, the authors propose Proto-ML, an integrated development environment that unifies prototype implementation, quality evaluation, and knowledge management into three cohesive modules within a single framework. Proto-ML enables structured documentation, multi-role collaboration, and the generation of reusable artifacts. User studies demonstrate that Proto-ML significantly enhances development efficiency while fostering a more transparent and reproducible machine learning development workflow.