Abstract
The uncontrolled growth of some cities and significant social inequalities were reflected in the population's housing conditions. Since housing projects must meet the needs of their inhabitants today and in the future, impacted by climate change, it is essential to search for more sustainable alternatives for social housing. To help decision-making on housing programs and incentives for sustainable building, this study aims to identify the main barriers and critical success factors of sustainable building projects. With this objective, a combination of qualitative data collection methods was applied, including the literature review and semi-structured interviews. The study's main results are the most significant barriers to sustainable social housing, such as lack of information, high cost, and lack of government incentives. Besides the critical success factors to guide sustainable social housing projects, the study highlights results such as the implementation of policies by government and professional bodies, monitoring compliance by the government, and commitment to sustainable delivery with the opinion of all stakeholders. Therefore, sustainable social housing projects can work as a tool to encourage social housing, strengthening the environmental pillar of sustainability. Furthermore, directly benefits the population with lower purchasing power, strengthening the economic and social pillars.

Abstract
This study investigates the evolutionary impact of applying artificial intelligence (AI) technology to foreign language education. By employing complex systems thinking, the relationship between foreign language education and AI technology is explored, and dynamic models are employed to analyze the evolutionary patterns of AI technology in foreign language education. Through model analysis and numerical simulations, the interactive effects between foreign language education and AI technology in different modes are revealed. The findings demonstrate that, under different coupling modes, foreign language education and AI technology can achieve self-organizing evolution. When the interaction coefficient between foreign language education and AI technology is appropriately set, AI technology exhibits emergent properties for foreign language education. Lastly, suggestions are presented to promote the sound development of foreign language education and AI technology.

Abstract
Abstract The article focuses on the consumption of goods used by consumers of different generations from 3 different countries: Albania, Polish and Portugal. The aim of the analysis was to identify respondents‘ indications concerning: (1) knowledge of the definition of the circular economy, (2) declared by respondents places of purchase of used products and (3) type of purchased products used by respondents. The analysis was conducted among 495 respondents from Albania, Polish and Portugal representing three generations (X, Y, Z). Correspondence analysis was used for statistical data analysis. Statistically significant differences in knowledge of the definition of the circular economy were shown between respondents from Albania, Polish and Portugal. It was also found that respondents‘ preferences regarding the place of purchase of second-hand goods are differentiated (at a statistically significant level) by nationality and year of birth (generation). The obtained results open the possibility of further research aimed at identifying different behaviors among these groups of consumers. The presented work, both in the cognitive and application part, can be a source of knowledge and popularization of research, as well as a source of inspiration for in-depth reflection and scientific discussion. The analyses presented in the publication may complement the existing research in the field of circular economy. Extending the survey to other EU countries can help define a strategy for policymakers, manufacturers and retailers to make greater use of circular economy solutions, while maintaining the viability of their operations.

Abstract
The main aim of the paper was to estimate the impact of older employees aged 55-74 on productivity compared with other age groups. The research (foundation of the paper) was conducted in 72 countries in the period 2000-2021. Countries were divided into three groups composed of 24 economies according to their GDP pc in 2021, i.e. the lowest GDP pc, the middle GDP pc, and the highest GDP pc countries. The study covered the five age and sex groups of employees: 25-34, 35-44, 45-54, 55-64, and 65-74. The Productive Capacities Index (PCI) built by UNCTAD and its selected categories were assumed to be dependent variables. The research results obtained for three groups of countries indicate that older employees had stronger positive impact on improving PCI than younger employees, especially on human capital development and private sector productivity. Our outcomes also suggest that older employees are better integrated than younger age groups with ICT tools that increase work productivity. To sum up, we can state that older employees can be a key factor in economic development due to their knowledge and experience, provided that others are willing to learn from them and they receive relevant organizational support.

Abstract
As the field of Machine Learning evolves, the number of available learning algorithms and their parameters continues to grow. On the one hand, this is positive as it allows for the finding of potentially more accurate models. On the other hand, however, it also makes the process of finding the right model more complex, given the number of possible configurations. Traditionally, data scientists rely on trial-and-error or brute force procedures, which are costly, or on their own intuition or expertise, which is hard to acquire. In this paper we propose an approach for algorithm recommendation based on meta-learning. The approach can be used in real-time to predict the best n algorithms (based on a selected performance metric) and their configuration, for a given ML problem. We evaluate it through cross-validation, and by comparing it against an Auto ML approach, in terms of accuracy and time. Results show that the proposed approach recommends algorithms that are similar to those of traditional approaches, in terms of performance, in just a fraction of the time.

Abstract
In the last years, the number of machine learning algorithms and their parameters has increased significantly. On the one hand, this increases the chances of finding better models. On the other hand, it increases the complexity of the task of training a model, as the search space expands significantly. As the size of datasets also grows, traditional approaches based on extensive search start to become prohibitively expensive in terms of computational resources and time, especially in data streaming scenarios. This paper describes an approach based on meta-learning that tackles two main challenges. The first is to predict key performance indicators of machine learning models. The second is to recommend the best algorithm/configuration for training a model for a given machine learning problem. When compared to a state-of-the-art method (AutoML), the proposed approach is up to 130x faster and only 4% worse in terms of average model quality. Hence, it is especially suited for scenarios in which models need to be updated regularly, such as in streaming scenarios with big data, in which some accuracy can be traded for a much shorter model training time.