New Trends for Feature Selection Applied in Data Mining

Dear Colleagues,

Feature selection has emerged as a critical preprocessing step in modern data mining, serving as the cornerstone for building efficient, interpretable, and robust machine learning models. As we navigate through the era of big data and artificial intelligence, the dimensionality of datasets continues to grow exponentially, presenting unprecedented challenges for data mining practitioners. This Special Issue aims to showcase cutting-edge research and innovative methodologies that address the evolving landscape of feature selection in data mining applications. By bringing together researchers, practitioners, and industry experts, we seek to explore novel theoretical frameworks, computational algorithms, and practical applications that push the boundaries of traditional feature selection paradigms and pave the way for more intelligent, adaptive, and scalable solutions.

The scope of this Special Issue encompasses a broad spectrum of research areas where feature selection plays a pivotal role in enhancing data mining performance. We welcome contributions that investigate theoretical foundations, algorithmic innovations, and practical implementations across diverse domains. The Issue particularly emphasizes the integration of feature selection with emerging technologies such as deep learning, federated learning, edge computing, and explainable artificial intelligence. We encourage submissions that address scalability challenges in high-dimensional data environments, explore multi-modal and multi-view feature selection scenarios, and develop solutions for real-time streaming data applications. Additionally, we seek research that bridges the gap between academic innovation and industrial deployment, fostering cross-disciplinary collaborations that advance both theoretical understanding and practical utility of feature selection methods in data mining contexts.

This Special Issue will focus on (but is not limited to) the following topics:

• Deep Learning-Enhanced Feature Selection: Integration of neural networks, attention mechanisms, and deep autoencoders with traditional feature selection paradigms to capture complex non-linear relationships and hierarchical feature representations in high-dimensional data spaces.
• Explainable AI and Interpretable Feature Selection: Development of transparent feature selection methods that provide human-understandable justifications for feature importance rankings, enabling domain experts to validate and trust automated selection decisions in critical applications such as healthcare, finance, and autonomous systems.
• Federated and Distributed Feature Selection: Novel approaches for collaborative feature selection across decentralized data sources while preserving privacy constraints, addressing communication efficiency, and maintaining model performance in federated learning environments.
• Multi-View and Multi-label Feature Selection: Advanced techniques for selecting informative features from heterogeneous data sources or multi-label data, including text, images, graphs, and temporal sequences, with applications in social media analysis, multimedia mining, and complex system monitoring.
• Real-Time and Streaming Feature Selection: Adaptive algorithms capable of processing high-velocity data streams, handling concept drift, and maintaining feature relevance in dynamic environments such as IoT networks, financial markets, and online recommendation systems.
• Hybrid and Ensemble Feature Selection: Combination of filter, wrapper, and embedded methods to leverage complementary strengths, improve selection stability, and enhance generalization performance across diverse data mining tasks and application domains.
• Big Data and Scalable Feature Selection: Efficient algorithms designed for massive datasets, including MapReduce-based approaches, GPU acceleration techniques, and approximate methods that balance computational complexity with selection quality in distributed computing environments.
• Domain-Specific Feature Selection Applications: Tailored methodologies for specialized fields including bioinformatics, medical diagnosis, cybersecurity, smart cities, and industrial IoT, addressing unique challenges and requirements specific to each domain.
• Automated Machine Learning (AutoML) for Feature Selection: Self-optimizing systems that automatically select appropriate feature selection techniques, hyperparameter tuning, and pipeline configurations based on data characteristics and performance objectives.
• Robust and Adaptive Feature Selection: Methods resilient to noisy data, missing values, and adversarial attacks, with capabilities for continuous learning and adaptation to evolving data distributions and environmental changes.

Dr. Ping Zhang
Dr. Yonghao Li
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• deep learning feature selection
• explainable AI
• federated learning
• multi-view data mining
• real-time streaming analytics
• high-dimensional data reduction
• hybrid ensemble methods
• AutoML feature engineering