ASSESSING THE REGIONAL LABOR MARKET BY USING DATA MINING METHODS: WAYS OF EFFECTIVE FUNCTIONING

Abstract

As a result of the uneven development of certain territories, it is more feasible and effective to tackle the practical issues of labor market regulation at the regional level. This ensures sufficient regulation of the system. Since it is necessary to properly account for the regional differences in practice, it is required that these issues be methodologically justified. Therefore, the aim of this paper is to investigate regional labor markets based on indicators of the socio-economic development of regions using the data mining methods. The current study has clustered regions of Ukraine on the basis of the level of their socio-economic development using data mining methods, in particular Kohonen maps and the k-means methods. One of the most critical stages in the assessment of Ukraine’s regions in terms of socio-economic development by using data mining methods is to determine the information base, criteria of evaluation, and a list of estimates. The data mining methods have gained much popularity in the assessing regional differentiation. The conducted analysis based on data mining methods included the use of the Deductor software, which includes the following analytical algorithms: neural networks, Kohonen’s self-organizing maps, autocorrelation and regression, associative rules, decision trees. For our study, we used the cluster analysis method based on Kohonen’s self-organizing maps as one of the most popular and frequently used methods for solving problems of the regional economy and assessing the differentiation of regions. In the context of our task, the result of cluster analysis is clusters of regions, united by indices of socioeconomic development. The main aspects of the socio-economic and demographic development of the regions are characterized by a set of statistical indicators related to four blocks of key factors: 1. Assessment of the demographic situation in a region. 2. Assessment of the social situation in a region. 3. Assessment of the economic situation in a region. 4. Assessment of the organizational environment in a region. The study, by no means, claims to detect all the dependences in the labor market related to all the above-mentioned factors. Based on public data, given in the statistical yearbook “Ukraine in Figures” (2020), by using mathematical methods (correlation-regression and cluster analysis), we obtained two groups of factors that characterize different aspects of the socio-economic and demographic development. The ranking of the regions by the level of extensive and intensive development shows that the development of the regions in Ukraine mainly proceeds in the extensive path of development. Almost all regions of Ukraine demonstrate a low level of intensive development. The integrated coefficient of intensive development for many territories is far from a maximum value; there are well distinguishable and huge discrepancies in the levels of the regions’ intensive development. Such a gap between the natural and human resource potentials, on the one hand, and the level of the development of economic activity and its territorial organization within the regions, on the other hand, leads to investment unattractiveness of some territories. Thus, the estimation of the country’s regions based on the level of their socio-economic development testifies to the dominance of extensive factors in the development of most regions in Ukraine. Common areas of the policy, conducted in the labor market, for all groups of regions are the measures to conduct an active policy (promoting self-employment and small businesses; the creation of new jobs; vocational training and retraining of unemployed people; public works; improvement of employment services, etc.).