Symbolic Methods of Machine Learning in Knowledge Discovery and Explainable Artificial Intelligence

Dear Colleagues,

Symbolic methods, also called interpretable or white-box methods, were one of the first methods developed within the machine learning area. These methods are still being developed and they find practical applications particularly in knowledge-discovery tasks. In predictive analytics complex approaches (complex AI/ML models) such as boosting, bagging and deep learning usually achieve better results than white-box methods. However, the explanation of a decision-making process of complex AI/ML models is difficult and, without some additional assumptions, often impossible. For this reason, such models are called black-boxes. The dynamic growth of XAI (Explainable Artificial Intelligence) has been recently stimulated by the necessity to explain decisions made by complex AI/ML systems. In this domain the most progressive development has been observed in local, so- called instance-level explanation (i.e., explanation of reasons for making specific decision for a given example). The global or dataset-level XAI still requires intensive research. Generally, the method of global explanation should help the user to understand how the AI/ML model makes decisions globally, for example, about the patterns of right and wrong decisions made by the AI/ML model. In this context, white-box based approximations of complex AI/ML models may play an important role. Specifically, in recent years research on approximation of decisions made by black-box models using white-box approaches has been done.

This Special Issue focuses on new methods of induction of interpretable AI/ML models (rules, trees, graphs, etc.) in data mining and knowledge discovery. The methods for concept learning, contrast set mining, action mining, regression and censored data analysis are welcome. The Special Issue covers also all proposals related to white-box based XAI dedicated to global explanation of decisions made by the complex AI/ML models.

You may choose our Joint Special Issue in Applied Sciences.

Prof. Dr. Marek Sikora
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.

• knowledge discovery
• white-box ML
• explainable artificial intelligence
• decision tree and rule induction
• rough sets