Clustering of business structures by the level of their digital maturity using two approaches: iterative and hierarchical

Abstract

Purpose. The aim of the article is experimental finding of the optimal number of clusters and their characteristic features for interpreted (understandable) segmentation of business structures by the level of digital maturity by several methods; comparing the results obtained by different methods and determining the most effective for a particular data analysis task. Methodology of research. Two clustering methods are used in the study: 1) using the Data Mining add-in for MS Excel spread sheets. Clustering capabilities in MS Excel are represented by iterative algorithms: k-means and Expectation-Maximization. For the reference, it was determined EM-algorithm; 2) using the functions of libraries for machine learning Python programming language. Findings. An experimental comparison of the use of two approaches to the clustering of respondents according to the results of online questionnaire using Google Forms service – hard and soft clustering, is conducted. Hard clustering was implemented using Python tools using the hierarchical agglomerative method, soft using the Data Mining add-in MS Excel and using the iterative EM method. A comparative analysis of the results obtained by the two methods showed that the agglomerative hierarchical clustering method is an effective method for solving the problem of clustering of mixed-type data obtained from the survey of respondents. Originality. An algorithm for solving the problem of respondents' clustering according to the results of online survey is proposed, including the stages of collection, preparation of data, obtaining the main results and development of future goals, which will help to solve the problem of processing and clustering of data of mixed type and to provide higher productivity of analytical data and analytical data. Practical value. This approach can be used to develop a Digital Transformation Index for domestic business structures and to measure their digital maturity, which will enhance their economic potential and therefore the country's economy. Key words: digital maturity; digital transformation; business structures; clustering methods; mixed-type data; surveys.