Interpretable Machine Learning and Statistical Modeling in High-Dimensional Data

Dear Colleagues,

In today's big data era, extracting valuable insights from high-dimensional data is a central pursuit for data scientists. Beyond traditional statistical models, machine learning algorithms represent a powerful, data-driven option for handling high-dimensional data. Many machine learning algorithms have been developed with an emphasis on improving prediction. However, in numerous application fields, notably in biomedicine, the interpretability of a model – such as its ability to lucidly delineate the connections between responses and predictors – is frequently a decisive element in its broad acceptance and utilization.

In this Special Issue, we invite researchers to contribute original research articles, comprehensive reviews, and insightful perspectives that explore innovative statistical models, machine learning algorithms, and methods of explainable artificial intelligence, all focusing on the challenges presented by high-dimensional data. This also encompasses tailored data visualization and computing techniques specific to high-dimensional data contexts. We are particularly interested in methodologies with bioinformatics, genetics, and biomedicine applications, but submissions from other disciplines are also warmly welcomed. The scope of potential topics is broad and not limited to these areas alone.

• Dimensional Reduction: This involves feature extraction and selection methods within high-dimensional data, with a special emphasis on approaches that retain interpretability following reduction. Specifically, it is desirable that the new features, obtained through dimensional reduction, maintain the ability to correspond to the original features and clarify their relationship with response variables.
• Statistical and Machine Learning Algorithms: This category invites contributions involving machine learning, data mining, or/and combined statistical models applied to high-dimensional data, where the focus is primarily on interpretability. This Special Issue places a higher value on the models' ability to explain and elucidate phenomena rather than solely on their predictive accuracy.
• Interpretable Machine Learning (iML) and Explainable Artificial Intelligence (XAI): We invite pioneering contributions in iML/XAI algorithms, which include intrinsic and post hoc approaches and those specific to a model or model-agnostic. This section aims to highlight innovations in making complex algorithms transparent and understandable.
• Higher-Order Interactions: This category focuses on issues related to higher-order interactions among features in high-dimensional data. This includes pure interactions (where there is feature interaction without any marginal effect) and impure interactions (featuring both interaction and marginal effects). Relevant topics might include dimensional reduction strategies that consider higher-order interactions, methods for detecting significant higher-order feature interactions that impact the response variable, visualization techniques for these complex interactions, and other topics related to higher-order interactions.
• Visualization: Effective visualization techniques and software for conveying critical information in high-dimensional data, highlighting important features, feature interactions, and correlations between features.
• Computing Techniques: This category focuses on exploring data parallel processing methodologies and the application of high-performance computing in modeling. We invite contributions highlighting advanced computing strategies to enhance model efficiency in handling complex data sets.
• Application Areas: Particularly welcoming topics in bioinformatics, genetics, and biomedicine. However, we are also open to innovative applications from various other fields. We seek contributions demonstrating the practical impact and implementation of high-dimensional data analysis across diverse scientific domains.

Dr. Yu-Chung Wei
Guest Editor

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 100 words) can be sent to the Editorial Office for announcement on this website.

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. Mathematics 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 2600 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.

• big data
• bioinformatics
• biostatistics
• computing technique
• data mining
• data visualization
• dimension reduction
• ensemble learning
• feature extraction
• high-dimensional data
• higher-order interactions
• logic
• machine learning
• statistical genetics
• statistical learning
• supervised learning
• unsupervised learning