Purpose: This study aims to examine the main aspects of innovation development regarding digital tools used in the tourism industry in a local area.Design/methodology/approach: A qualitative approach and the descriptive analysis technique was employed to analyse completed semi-structured expert survey questionnaires created in Google Forms. The aim is to examine the opportunities and potential for digital innovation in tourism in the Baltic region. Specialists representing tourism and hotel businesses, the guide profession as well as the disciplines of business and economics were selected as experts.Findings: The experts indicated that digital marketing was very important for tourism development, and it is still considered a new field of science with huge potential. The research results showed that when introducing/developing digital innovations in a local area, tourism companies were faced with low-quality or no data analytics at all. In addition, long-term maintenance of digital tools was not considered, and the investment outweighed the benefits of implementing the technologies.Originality/value: The study contributes to the development of digital innovation and the identification of problems in a local area.

The objective of this paper is to examine the efficiency of the public sector when it comes to labor market policy (LMP) in European Union (EU) countries. The primary aim is to provide a comprehensive evaluation of public services connected with the LMP from the viewpoint of the efficiency of public expenditure allocated for that purpose. The turning point for the European labor market was marked by the introduction of lockdown in 2020 due to the COVID–19 pandemic. The article’s overall aim is to evaluate the efficiency of EU labor markets before and during the pandemic and to ascertain the extent to which the volume of public services (their financial allocations) affects selected labor market indicators. The efficiency of EU countries in 2019 and 2020 was evaluated by applying the Data Envelopment Analysis (DEA) method. The BCC (Banker, Charnes, Cooper) model was used in this study. The output‑oriented BCC model was selected as it seems adaptable to the specifics of the LMP. Microsoft Excel Solver was used for solving the DEA model, which is applied to working out linear optimization models. The efficiency of public expenditure was evaluated using data from Eurostat. A total of seven variables were used for the analysis; they comprised five variables as inputs (government expenditure on services by selected function and two uncontrolled inputs represented by job vacancy rate and GDP per capita) and two variables as outputs of selected labor market indicators (employment rate and unemployment rate ). The cross‑country comparison was conducted in two dimensions – among individual EU countries and among a group of countries that represent different public administration traditions (i.e., Anglo‑Saxon, Continental European, South European, Scandinavian, Eastern European, and South‑Eastern European). The study demonstrates that even though the Southern countries achieved the worst results in terms of labor market indicators, it is in contrast to the efficiency approach using DEA, where countries such as Portugal and Greece achieved full efficiency for both 2019 and 2020. The efficiency approach of the LMP for 2019 and 2020 indicates that there is a very good situation in Central and Eastern European countries. Full efficiency is also achieved by Cyprus, Malta, Sweden and Germany. The least efficient LMP is found in Belgium. During the pandemic, most of the analyzed groups of countries improved their efficiency scores of public expenditure on LMP. The exceptions were groups of countries with the South European tradition and the Scandinavian tradition, where the measure of efficiency decreased slightly. This may demonstrate the effectiveness of the implemented public policy to counteract the negative effects of COVID–19 on the labor market. The value added of this paper comes from the demonstration that the proposed methodology, using the DEA method, can be used to measure the efficiency of LMP at micro‑ and mezzo‑levels and to make cross‑country comparisons.

The purpose of the study is to substantiate the theoretical and methodological provisions and develop practical recommendations for the development of green entrepreneurship for the post-war recovery of Ukraine based on the experience of Latvia. The scientific novelty of the study lies in solving an important scientific and applied task - further development of theoretical aspects of green entrepreneurship, substantiation of the relationship between eco-innovation, green growth strategies and environmental policy in the context of sustainable development. The study uses general scientific and special methods: scientific abstraction, deduction, analysis and synthesis, systemic and critical analysis, structural and logical, ascent from the abstract to the concrete, and statistical analysis. It is proved that the Latvian business ecosystem is adapted for highly efficient investments and innovations in the field of green technologies. The study identified guidelines for environmental policy that will promote the development of green entrepreneurship, increase the density of green infrastructure, extend the life cycle of high value-added products manufactured in the country, reduce waste and encourage recycling. The importance of studying the experience of Latvia in strengthening the strategic framework for sustainable development and green growth for the post-war recovery of Ukraine is substantiated. It is determined that improving the quality of environmental infrastructure and services, as well as investing in innovation potential, will ensure the decoupling of economic growth from resource use and environmental depletion and accelerate the transition to a closed-loop economy. The authors propose the components of a «green» post-war recovery of Ukraine in the context of its further European integration.

Publisher: Vilnius University Institute of Data Science and Digital Technologies, Journal: Informatica, Title: Stochastic Optimization Algorithms for Support Vector Machines Classification, Authors: Vaida Bartkutė-Norkūnienė , In this paper, we consider the problem of semi-supervised binary classification by Support Vector Machines (SVM). This problem is explored as an unconstrained and non-smooth optimization task when part of the available data is unlabelled. We apply non-smooth optimization techniques to classification where the objective function considered is non-convex and non-differentiable and so difficult to minimize. We explore and compare the properties of Simulated Annealing and of Simultaneous Perturbation Stochastic Approximation (SPSA) algorithms (SPSA with the Lipschitz Perturbation Operator, SPSA with the Uniform Perturbation Operator, Standard Finite Difference Approximation) for semi-supervised SVM classification. Numerical results are given, obtained by running the proposed methods on several standard test problems drawn from the binary classification literature. The performance of the classifiers were evaluated by analyzing Receiver Operating Characteristics (ROC).

The research aims at the identification of factors influencing mentoring services with a particular emphasis on the country of origin of the mentor and the mentee. The quantitative research was conducted in four Central and Eastern European countries, i.e., Bulgaria, Poland, Latvia and Lithuania and one Southern European country, Italy. The implemented methodology covered Exploratory Factor Analysis (EFA) and reliability analysis, which were performed to identify factors influencing the process of communication in the mentor–mentee relationship. Moreover, to diagnose statistically significant differences between individual countries in terms of factors influencing the effectiveness of communication, the Kruskal–Wallis H Test and the Mann–Whitney U Test, as well as pair-wise comparisons, were used. Factors influencing communication in the mentor–mentee relationship are mentor traits, mentor’s personal background, mentor’s professional background, non-verbal communication channels, communication barriers, written communication channels, online communication, quality of content and the ability of content processing by the mentee. There are differences in the perception of individual factors in the analysed countries. So far, no cross-country comparison has been conducted of factors influencing mentoring services. As a direction for future research, more detailed research can be recommended concerning factors of the mentoring process in such countries as Lithuania, Latvia and Italy by developing separate models (or EFA) for mentors and mentees.