Using the Clustering and Visualization Data Mining Algorithms in Digital Currencies

This paper demonstrates the potential application of artificial intelligence - specifically, clustering algorithms and related data visualization techniques - in the context of digitalization of public finance. The analysis is conducted of the possibility of introducing central bank digital currencies as a new instrument of monetary policy. The research proposes a methodological approach based on the adaptation of the Cross-Industry Standard Process for Data Mining framework to cluster analysis, specifically for clustering countries to evaluate their readiness for Central bank digital currencies implementation. The data preprocessing stage included handling missing values, outlier detection, feature scaling, and correlation analysis. Subsequently, clustering algorithms requiring a predefined number of clusters were employed, including Hierarchical Clustering, K-Means, K-Medoids, Spectral Clustering, and Balanced Iterative Reducing and Clustering using Hierarchies (with the number of clusters determined using the elbow method, dendrogram analysis, and theoretical justification). Additionally, clustering algorithms that do not require a predefined number of clusters were applied, such as Affinity Propagation, Mean Shift, Density-Based Spatial Clustering of Applications with Noise, Hierarchical Density-Based Spatial Clustering of Applications with Noise. To evaluate the clustering results, the following quality metrics were calculated: Silhouette Score, Davies - Bouldin Index, and Calinski - Harabasz Index. The application of cluster analysis revealed the possibility of identifying countries with similar characteristics and conditions and grouping them in terms of their readiness to implement Central bank digital currencies. The theoretical contribution lies in substantiating the applicability of clustering and visualization techniques for the analysis of public finance, including the identification of cross-country similarities and differences in the context of Central bank digital currencies deployment. The practical significance of the study is in the development of an approach for assessing the conditions necessary for the successful implementation of Central bank digital currencies, which may assist central banks in designing optimal deployment strategies.