Use Of Cluster Analysis Research Articles

Value orientations and attitudes towards money among Russian and Chinese students as characteristics of the economic behavior of young people

Introduction. In the global economy, young people from the perspective of public administration are considered as intellectual capital. The purpose of this article is to conduct a comparative analysis of value orientations and attitudes towards money as indicators of economic behavior among Russian and Chinese students. Methods. The study used the following methods on a sample of 88 Russian and Chinese students: 1) Questionnaire by B. Klontz and T. Klontz “Monetary Attitudes” adapted by Bayazitova, T.A. Lapshova; 2) Portrait Value Questionnaire by S. Schwartz (PVQ-RR) adapted by T.P. Butenko, D.S. Sedova, A.S. Lipatova); 3) Questionnaire “Moral assessment of money” E.I. Gorbacheva and A.B. Kupreichenko. Statistical analysis of the data used the Mann-Whitney U test, Spearman's rank correlation coefficient and cluster analysis. Research results. In the process of conducting empirical analysis, significant differences were identified between Russian and Chinese students in the manifestations of value orientations such as: 1) reputation (0.036; p ≤ 0.05); 2) personal safety (0.04; p ≤ 0.05); 3) conformity - rules (0.026; p ≤ 0.05); 4) modesty (0.004; p ≤ 0.01). Russian students value their reputation and personal safety more; on the money avoidance scale, Chinese students have a significantly higher indicator compared to Russian students (0.026, p ≤ 0.05). As a result of the correlation analysis, significant relationships between the scales were found among Russian students: (p ≤ 0.05): 1) avoidance of money - hedonism (r = -0.300), modesty (r = 0.298), benevolence (r = -0.353); 2) money as status – modesty (r = -0.333); 3) worship of money - independence of thought (r = 0.375), independence of action (r = 0.331), hedonism (r = 0.341), power - dominance (r = 0.351), security - personal (r = 0.336), benevolence - duty ( r = 0.330). Among Chinese students, significant relationships were found for the following indicators: In the group of Chinese students, the following relationships were identified (p ≤ 0.05): 1) money as status - power - resources (r = 0.312), traditionalism (r = 0.302); 2) worship of money - independence of thought (r = 0.310), conformity - rules (r = 0.320), conformity - interpersonal (r = 0.372), modesty (r = 0.318); 3) vigilance regarding money - achievement (r = 0.315), power - resources (r = 0.309), conformity - rules (r = 0.304). The use of cluster analysis made it possible to identify positive and moral-conflict types of attitudes towards money among Russian students, and among Chinese students - morally uncertain and compromising-positive types of attitude towards money. Conclusion. As a result of the study, recommendations were proposed for the development and correction of indicators of economic behavior among young people: 1) introduction into educational practice in higher education of methods for developing students' knowledge in the field of financial literacy; 2) development of civic identity and patriotic feelings, which act as psychological support for students on the path to professional development and financial success; 3) developing students’ attitude towards money as a result of personal and professional self-realization; 4) providing information and education to foreign students in the field of economic behavior in Russia in order to successfully adapt to a foreign cultural environment.

Динамика углеродного статуса почвы на ранней стадии роста растений ячменя под влиянием средств биоактивации почвы

The article presents an approach to data processing that extends critical indices of plant structural analysis by new characteristics of vegetation indices. The ability to detect differences in plant genotype usingvegetation indices was confirmed using common millet (Panicum miliaceum) as an example. The use of cluster analysis to group the tested breeding lines into clusters with optimal variety characteristics for further breeding work was demonstrated. In 2021…2023, practical results were obtained in the form of transferred new varieties of winter soft wheat (Triticum aestivum L.), standing out or having greater productivity in clusters with optimal parameters: Aist Orlovskiy, Zusha, Pamyati Katkova and Akvamarin. The conclusion is drawn on the promising use in the breeding process of groups with reduced tillering based on the variety sample Timiryazevskaya Odnostebel’naya. Promising breeding lines with highly productive ear based on the variety Ferruginum and new forms of Ferruginum and Milturum with high values of vegetative indices in combination with a complex of indicators characterizing yield and grain quality were revealed.

Understanding the Use of Cluster Analysis in Business

Understanding the Use of Cluster Analysis in Business

The use of cluster analysis to assess thermal comfort in university classrooms

Nowadays, providing health, well-being, productivity and energy efficiency to users inside buildings is essential. Applying these aspects aligned with sustainability becomes necessary to reduce the use of heating, ventilation, and air-conditioning (HVAC) systems. These systems are currently used to provide better thermal conditions to the occupants, who spend around 80% of their time indoors. The actual thermal conditions can be affected by several factors, such as the climatic type of the region, orientation, size, building type, and energy levels, among others. To assess thermal conditions inside buildings, several thermal comfort models have been developed over the years. However, the Predicted Mean Vote (PMV) created by Fanger is still the most common model to assess thermal comfort indoors. In this context, this research aimed to analyze thermal comfort conditions in university classrooms in Southern Brazil. By collecting the environmental and personal variables of thermal comfort and the mean thermal sensation of students through measurements and questionnaires, a total of 519 responses were obtained during the Brazilian autumn. A statistical cluster analysis was performed to classify individuals according to their sensations. Differences between genders were verified and changing indoor temperatures lower in winter would therefore save HVAC energy without impacting occupant comfort.

Comparison of Morphometric Traits of Pecan Cultivars

Pecan is a major horticultural nut tree that originated in North America and, in Turkey, it is generally cultivated in the Aegean and Mediterranean regions. In this study, morphometric traits of twenty-one cultivars grown under equal ecological conditions in the Southeastern region of Turkey were determined. In addition, a multivariate analysis was performed on morphometric traits for determining suitable cultivars that show the best performance in these ecological conditions. The nut weight ranged from 3.66 g for ‘Cherokee’ to 10.35 g for the ‘Shoshoni’ cultivar. The ‘Choctaw’ had the highest kernel weight and kernel ratio ranging from 5.63 g to 58.60%. While the ‘Cherokee’ had the lowest ranging from 0.91 g to 22.84 g, respectively. A significant positive correlation between nut weight and kernel ratio was determined. The kernel weight was also highly correlated with nut weight and height. The use of clustering analysis according to Ward’s method allowed the establishment of relationships between pecan cultivars by separating them into four main clusters. Also, these results were supported by principal component analysis, and ‘Big Z’ and ‘Ideal’ cultivars were determined to be the closest cultivars. Results showed that there are quantitative differences regarding the genetic property of pecan cultivars and more than half of the examined cultivars showed a good performance in terms of nut quality criteria that are crucial in marketing. Furthermore, the multivariate analysis was effective in the differentiation of pecan cultivars and it has been evaluated that it can be used successfully in future studies.

Кластерный анализ демографических процессов в Казахстане на современном этапе

The article is devoted to the analysis of demographic processes using cluster analysis. The application of the method of cluster analysis for the selection of territorial groups in the regional context is proposed. The relevance of the application of the method is substantiated by the fact that today the average statistical data do not allow considering the characteristics of the population in a complete picture. The review of the results of the application of cluster analysis, which reflected the unfavorable characteristics of the population, describes the possible consequences as a result of the impact of these unfavorable characteristics. The need to classify territories is indicated without considering the regional centers of attraction of the population — regional centers, large cities and the capital. This approach makes it possible to note and trace the problems of excessive concentration of the population of Kazakhstan in cities — regional centers of attraction and exposure of nearby and rural areas. Further application of the method of territorial classification with the use of cluster analysis in the study of demographic processes of the population is proposed. The results of the cluster analysis demonstrate a variety of characteristics data for Kazakhstan in connection with the methodology for allocating territories. The study results designates a more appropriate assessment of the characteristics of the population, reveals the problems of the demographic characteristics of the population.

Using the Sectoral Structure of the Economy to Select Competing Regions (on the Example of the Amur Region)

One of the stages of the statistical study of the competitiveness of a region is the selection of competing regions.Purpose of the study. The purpose of the article is to form a statistical set of regions-competitors based on the sectoral structure of the economy.Materials and methods. As research methods in this article, the method of the main array, factorial, cluster methods, statistical methods are chosen. The statistical data of Rosstat were used for the study. To perform the calculations, the GVA was considered in the structure of Russian National Classifier of Types of Economic Activity2 for 2019. Results. With the help of factor analysis, 19 types of economic activity of the regions were grouped according to similarities and differences. As a result, six factors were formed, each of which collected dependent types of economic activity. The use of cluster analysis made it possible to form groups of regions with a similar sectoral structure of the economy. The study involved 85 regions of the Russian Federation. Cluster analysis made it possible to solve the methodological problem of determining the boundaries of GVA intervals for certain types of economic activity in the selection of competing regions.The paper shows that for the Amur Region, nine regions of the Russian Federation should be considered as competing regions. The regions of this cluster are united by a high share of gross value added by the types of activity “Transportation and storage”, “Public administration”, “Trade”. At the same time, competitors are regions from different federal districts: 70% of the regions of the Far Eastern Federal District, 20% of the Southern Federal District, 10% of the Siberian Federal District. The main results of the study are the following: 1) a high variation of the regions of the Russian Federation in 2019 was revealed by the type of economic activity “Mining” and “Manufacturing”; 2) a grouping of 19 types of economic activity of the regions was carried out using the factor analysis method; 3) a cluster analysis of the regions of the Russian Federation was carried out according to the sectoral structure of gross value added for 2019; five clusters were received. Conclusion. This paper shows that the selection of competing regions must be carried out using the sectoral structure of the region’s economy. Consideration of the region’ specialization is an important requirement of the selection methodology. The advantage of the author’s methodology is its universality, objectivity and reflection of the specialization of the region. As a direction for further research, one should consider determining the specialization of regions using localization coefficients and, on its basis, the formation of a statistical set of competing regions. The presented sample of regions is necessary for assessing their competitiveness.

Analysis of Spatial Distribution Processes for Forest Fires Near the Railway Infrastructure Using Clustering: Case Study

Almost all the states of the world community have territories along which railroad tracks pass. In a number of states, these railway lines pass through forests. As known, railway facilities are a source of anthropogenic load on forested areas. This causes the occurrence of forest fires near railway infrastructure. For medium-term and long-term prediction of forest fire danger, it is often necessary to process data on the forest fire retrospective. In this paper, the processing of statistical data on forest fires has been carried out by using cluster analysis. The purpose of the study is cluster analysis and processing of statistical data on forest fires near the infrastructure facilities of Russian Railways in the context of assessing and predicting forest fire danger. Main objectives: 1) collecting data from forestries; 2) preparing samples; 3) classifying two sets of samples; 4) analysing clusters obtained. In the work, data processing has been carried out for a typical territory of the Republic of Buryatia (Russian Federation), along which the railway passes. Hierarchical clustering algorithms have been used for data processing. The initial information for data processing has included meteorological parameters, forest stands description, statistical data on forest fires and data on the location of railway infrastructure facilities. By using the QGIS software, digital maps have been built showing the spatial distribution of forest fire clusters near the railway infrastructure, forest conditions and typical meteorological conditions. Based on the results of the work, conclusions and proposals have been formulated on the use of cluster analysis in predicting forest fire danger. An analysis of the spatial distribution of forest fire clusters over the territory of the Zaigraevsky district of the Republic of Buryatia shows that a large number of forest fire clusters are located near the railway infrastructure. In fact, railway facilities are a serious source of anthropogenic load. In the future, it is necessary to define the so-called buffer zones near railway infrastructure facilities, whose forested areas require careful attention when predicting forest fire danger. The results of cluster analysis confirm the data of descriptive analysis for a number of positions. For example, forest fires occur near railway infrastructure, which is explained by the presence of attractive points in forest areas due to logging, hunting, picking berries and mushrooms, as well as recreational pressure. As a result of the descriptive analysis, reasonable assumptions have been made and explanations have been given for the presence of local maxima of forest fires for the period 2016-2020, as well as during the fire danger season. The article also presents a plan for further research in this direction.

Assessment of the Impact of Scientific and Technical Activities on the Economic Growth of World Countries

At present, scientific and technical activities (STA) play a fundamental role in the creation of new knowledge necessary for the realisation of the goals of sustainable economic growth. The purpose of the research is to assess STA in countries of the world and Ukraine as well as to analyse the impact of STA on their sustainable economic growth. For the assessment of STA of world countries, we have developed a composite indicator (Index of STA–ISTA) based on two sub-indices (Potential of STA and Results of STA) and a set of partial indicators. The study is carried out using a theoretical and methodological approach that includes five research stages: calculating the ISTA of world countries with their subsequent ranking; assessing the impact of the ISTA on the economic growth of world countries; evaluating the correlations between the ISTA, the Global Innovation Index (GII), and the Human Development Index (HDI); breaking down world countries into groups with the help of cluster analysis, using data on the ISTA, GII, HDI; assessing and analysing the importance of environment-related technologies and innovations in ensuring the economic growth of countries of the world. According to the results of the study, the importance of STA in ensuring sustainable economic growth of countries of the world was confirmed. The use of cluster analysis to group countries of the world in terms of the level of STA made it possible to identify three clusters of countries: leaders, potential leaders, and catching-up countries. It has been proven that in order for a country to achieve leadership in terms of STA, it is important to harmoniously combine efforts to support and develop its scientific potential as well as to create principles for the implementation of scientific achievements in innovation activities to ensure its economic growth and improve the living standards of its population. Besides, the analysis allowed proving the existence of a moderately positive impact of the development of environment-related technologies and innovations on the acceleration of innovative development, improvement of the quality of life of the population, and the growth of GDP per capita.

The Use of Cluster Analysis to Assess the Threats of Poverty or Social Exclusion in EU Countries: The Case of People with Disabilities Compared to People without Disabilities

The main research objective was to assess the range of threats related to poverty or social exclusion in EU countries among people with disabilities as compared to people without disabilities. The research used the available poverty determinants, including the percentage of low work intensity, the percentage of low income, the percentage of material deprivation, the poverty risk rate, the percentage of severe disability, living conditions and population income, and the overall risk of household poverty or exclusion, which are available in Eurostat databases. The data used in the research relates to 2018 and was published in 2021. The research used cluster analysis, more specifically one of the agglomeration clustering methods, i.e., Ward’s method. Separate cluster analysis using Ward’s method was carried out for people with disabilities and for people without disabilities. The analysis identifies two clusters among people with disabilities and two clusters among people without disabilities. In the group of people with disabilities, cluster 2 includes 19 countries with a higher risk of poverty or social exclusion, while cluster 1 includes eight countries with a lower risk of poverty or social exclusion. In turn, in the group of people without disabilities, cluster 2 includes nine countries with a higher risk of poverty or social exclusion, while cluster 1 includes 18 countries with a lower risk of poverty or social exclusion.

USE OF FORECASTING AND MODELING METHODS FOR OPTIMIZATION OF HOTEL, RESTAURANT AND TOURIST BUSINESS DEVELOPMENT MANAGEMENT

The purpose of the article is to determine directions for the use of forecasting and modeling methods to optimize management of the development of the hotel and restaurant business. Types and methods of forecasting are listed. Classifications of forecasting methods were studied and their main groups were listed. The advantages and disadvantages of forecasting and modeling methods that are most commonly used in practice are analyzed. It was determined that the advantage of expert forecasting methods is relative simplicity for any situations, including in conditions of incomplete information. It is noted that simple methods of extrapolation can be used at the initial stage of forecasting to determine trends in changes in indicators. Complex methods of extrapolation can be used in the development of short-term forecasts. Smoothing methods are auxiliary in forecasting, which are used in combination with the main ones. It is noted that modeling methods can be used to determine the influence of various factors on the development of economic processes in the hotel and restaurant business. The advantages and disadvantages of the method of the annual average of the index are analyzed. The content of the forecasting methods themselves significantly depend on the terms of the forecasts. It was determined that when forecasting the development of hotel and restaurant business enterprises, it is most appropriate to use combined methods that combine economic-mathematical modeling and development scenarios inherent in intuitive methods. Recommendations for the use of cluster analysis of hotel and restaurant business enterprises are presented and the results of this analysis are determined. The stages of cluster analysis of a group of hotels and its results are described. It is noted that the possibilities of practical use of this or that forecasting method are determined by the purpose of the forecasting results, the features of the object, and the availability of the necessary information. Keywords: forecasting methods, modeling methods, economic and mathematical analysis, cluster analysis.

Standard Precaution Knowledge and Behavioral Intentions Among Students in the Healthcare Field: A Cross-Sectional Study.

Infectious diseases are a major threat to healthcare workers and patients alike. Standard precautions (SPs) are a pivotal element in controlling their spread. However, worldwide reported compliance with SP guidelines is suboptimal among workers and students in the healthcare field. The objective of this study was to identify the knowledge and behavioral intent patterns of students in the healthcare field as well as the relationship of these patterns to the students' study program, clinical exposure, and attitudes. This cross-sectional study employed a web-based structured questionnaire. Participants were students of nursing, emergency medicine for paramedics, physiotherapy, and medicine studying at a public university in Israel. The questionnaire assessed SP-related knowledge and behavioral intent, perceptions of personal responsibility, and the presence of mentors as role models. The K -means clustering method was used. The 259 students (33% response rate) who completed the questionnaire were enrolled as participants and divided into two knowledge and behavioral intent clusters. Cluster 1 ( n = 156) had a higher proportion of nursing students and a lower proportion of medical students than Cluster 2 ( n = 103). Emergency medicine for paramedics and physiotherapy students were more evenly distributed between the clusters. Participants who were classified in Cluster 1 reported higher intent with regard to proper hand hygiene, use of personal protective equipment, and decontamination practices. In multivariable analysis, only three variables predicted belonging to Cluster 1: nursing students, perception of their mentors as role models, and perception of having higher personal responsibility for microbiological safety. We observed clear patterns of knowledge and behavior among the students in the healthcare field examined in this study. These patterns were influenced by profession, role mentoring, and sense of responsibility. This use of cluster analysis may contribute to the development of better SP educational endeavors. The patterns identified highlight the need for improved training and the importance of role modeling for healthcare students in all areas of specialization.

The “affected” pharmacist and the “business as usual” pharmacist: Exploring the experiences of pharmacists during COVID-19 through cluster analysis

BackgroundThe global coronavirus disease 2019 (COVID-19) pandemic has necessitated considerable changes in the delivery of pharmacy services, with pharmacists experiencing increasing demands and a high rate of burnout. The ability to categorize pharmacists based on their burnout risk and associated factors could be used to tailor burnout interventions. ObjectiveThis study aimed to identify subgroups (profiles) of pharmacists and use these profiles to describe interventions tailored to improve pharmacist’s well-being. MethodsA survey was disseminated to pharmacists working in Australia during April and June 2020. The survey measured demographics, burnout, and psychosocial factors associated with working during COVID-19. A two-step cluster analysis was used to categorize pharmacists based on burnout and other variables. ResultsA total of 647 survey responses contained data that were used for analysis. Participants were mostly female (75.7%) and working full time (65.2%). The final cluster analysis yielded an acceptable two-cluster model describing 2 very different pharmacist experiences, using 10 variables. Cluster 2 (representing 53.1% of participants) describes the “affected” pharmacist, who has a high degree of burnout, works in community pharmacy, experiences incivility, is less likely to report sufficient precautionary measures in their workplace, and has had an increase in workload and overtime. In contrast, cluster 1 (representing 46.9% of participants) describes the profile of a “business as usual” hospital pharmacist with the opposite experiences.Interventions focused on the “affected” pharmacist such as financial support to employ specialized staff and equitable access to personal protective equipment should be available to community pharmacists, to reduce the risk to these frontline workers. ConclusionThe use of cluster analysis has identified 2 distinct profiles of pharmacists working during COVID-19. The “affected” pharmacist warrants targeted interventions to address the high burnout experienced in this group.

The Co-construction of News Values on News Magazine Covers: A Corpus-assisted Multimodal Discourse Analysis (CAMDA)

ABSTRACT Extant research from various perspectives has provided valuable insights into what news values are more frequently foregrounded across news types, topics, and outlets; and how they are constructed through verbal and visual resources. However, no research so far has examined how specific news values may be frequently co-constructed to package an individual news story and make it more appealing to potential readers. This research adopts the discursive news values analysis (DNVA) approach to examine the co-construction of news values in the multimodal discourse of TIME magazine covers (n = 300). To identify such co-construction, this study adopts cluster analysis, which has yet been deployed to study news values. The use of cluster analysis has identified six different combinations of news values repeatedly been observed together in clusters of covers. These combinations may represent what the magazine perceives as most attractive to the target audience. This study has also revealed intersemiotic relations in news values construction. While some news values are often reinforced across the verbal (i.e., cover line) and visual (i.e., cover image) semiotic modes, others rely more on the verbal mode. This study crosses disciplinary boundaries by bringing corpus linguistics techniques into journalism studies to understand a previously under-explored news genre.

On the relationship between anosmia/dysgeusia and nonspecific immunoprophylaxis COVID-19

The issues of non-specific immunoprophylaxis in the patients with new coronavirus infection have been raised since the WHO announced the COVID-19 pandemic. According to numerous studies, the new coronavirus infection is accompanied by manifestations of anosmia and dysgeusia. The purpose of this study was to perform a retrospective study of the relationships between nonspecific immunoprophylaxis of COVID-19 and conditions of anosmia/dysgeusia in the persons who underwent this infection in Sverdlovsk region over 2020.
 We have studied clinical and laboratory data of 84 employees at the general medical institution providing emergency pediatric care. All participants suffered a single infection caused by SARS-CoV-2. Etiological diagnostics included mandatory virological PCR testing of biological samples. The concentration of antibodies to the virus was determined using a set of reagents for SARS-CoV-2-IgG detection (JSC Vector-Best, Novosibirsk, Russia). A sufficient part of this group (n = 41; 48.8% of total sample) reported self-administration of non-specific immunoprophylactic therapy of COVID-19 (without consulting a doctor) since the beginning of coronavirus pandemics. To this purpose, cholecalciferol, riamilovir, human recombinant IFNa-2b, umifenovir hydrochloride monohydrate, ascorbic acid, zinc acetate were used. In 58 cases (69% of total), clinical course of infection was complicated by loss of taste and sense of smell (group No. 1), in 26 people (31.0%), no changes in taste and sense of smell were detected (group No. 2). Statistical evaluation of the data was performed using STATISTICA v.12.5.192.5 package (StatSoft, Inc., USA). Along with basic statistics, cluster analysis was performed.
 The use of cluster analysis allowed us to establish that there is a connection between the incidence of anosmia/dysgeusia and usage of distinct immunoprophylactic agents. In particular, the use of human recombinant IFNa-2b is protective in terms of reducing the number complications in COVID-19, especially, anosmia/dysgeusia caused by the new coronavirus. The use of this drug was associated with 8.5-fold decrease in the number of complications by (p = 0.03). Moreover, invasive intranasal usage of interferon, was associated with decreased hospitalization terms by 14.3% (p = 0.01); lower volume of lung tissue damage (p = 0.03), higher concentrations of IgG to SARS-CoV-2 at 2 months after reconvalescence (p = 0.001).
 For the first time, data were presented on the relationship between usage of immunoprophylactic agents and manifestations of anosmia/dysgeusia in COVID-19. The protective role of human recombinant IFNa-2b has been shown in terms of reduced incidence of the disease complications, e.g., anosmia/dysgeusia, degree of lung damage, as well as development of an antiviral humoral immune response. The data obtained could be used to substantiate clinical recommendations for prevention of new outbreaks of coronavirus infection. The limitation of the obtained results is small number of cases, thus requiring additional studies in a wider sample of respondents.

Multi-level stratification of territories for waste composition analysis.

The fundamental knowledge at all levels of decision-making related to waste management is the quantity and composition of waste. Many articles deal with methods for estimating the composition of municipal waste, but most details are given as to how many categories should be chosen and what technical procedure should be followed. In order to obtain a broader view and a reasonable evaluation of the results, it is necessary to select the areas where the analyzes will be performed effectively. Current approaches have insufficiently addressed this issue at the regional and national levels. This paper presents a method that uses multi-level stratification to divide municipalities into similar groups to reduce the number of observations needed to obtain an estimate of the composition of waste in a selected area (region or state level). The method combines expert knowledge with statistical considerations and makes use of cluster analysis. Socio-economic and waste-related parameters are used within the individual steps. Regarding the available financial resources and the required accuracy of the results, the municipalities in which the analyzes should take place are selected. These representative municipalities represent other municipalities in the created groups, and thanks to them, it is possible to estimate the composition of waste in any municipality, region, or larger territory. Waste analysis planning is an essential procedure for waste management, but the respective costs represent a crucial factor at the national level and even more for individual municipalities. Estimating waste composition impacts the transition to sustainable waste management and is thus a key element for further development in this sector. The presented method demonstrates the selection of 10 representative municipalities from the Czech Republic, but an arbitrary number can be set respecting the financial resources. Estimating the composition of the mixed municipal waste for the Czech Republic should cost around 72,000 euros for ten representatives with different distribution of dwelling types. The method is described in general and can be applied to any territory/country in the world, considering local conditions and possibilities.

Research on the success of unsupervised learning algorithms in indoor location prediction

With location-based smart applications, the flow of life can be facilitated and support can be provided in case of security and emergency situations. Indoor location detection provides various conveniences in complex structures such as hospitals, schools, shopping centers, etc. Indoor location detection studies are carried out by using data related to location and signal and machine learning methods. Machine learning has become frequently used as a solution method in this field, as in many other fields. When the studies in the literature are examined, it is seen that the studies are mainly focused on producing solutions with supervised machine learning algorithms. Unsupervised algorithms are frequently used to determine the labels of data groups that do not have labels. In this direction, it can be seen as the first step in labeling the data collected in indoor positioning studies and then using it for training predictive models to be developed with supervised learning methods. For this reason, the results to be obtained regarding the success and usefulness of cluster analysis will constitute an important basis for further studies. In this study, it is aimed to examine the success of unsupervised learning, in other words, clustering algorithms. The Wireless Indoor Localization Data Set and well-known k-Means and Fuzzy c-Means algorithms have been used with different distance measure. The obtained methods performances have been evaluated with internal and external indices. The results show that the clustering algorithms can cluster correctly data points in the range of 93-95% according to the accuracy and F measure value. Although performances indicators are very close to each other according to the internal indexes, it can be stated that the model obtained using the Manhattan distance measure and the k-Means algorithm has higher performance in terms of clustering success.

A Cluster Analysis of Population Based Cancer Registry in Brunei Darussalam: An Exploratory

Machine learning techniques have been mostly applied in gene expression cancer data. Socio-demographic data available in cancer registries could be explored, to get further insight into relationships between cancer types and their contributing factors. Moreover, less attention has been paid to analyse the mixed demographic data (numeric and categorical) from cancer registries and its association to the cancer types. The aim of this study is to identify subgroups of patients, having similar demographics characteristics, from the population based cancer registry in Brunei Darussalam and examine the prevalent cancer types in these subgroups. Four clustering algorithms are explored in the cluster analysis of Brunei Darussalam Cancer Registry; Two-step, Partitional Around Medoid, Agglomerative Hierarchical and Model-based. Gower distance was used for measuring similarity for mixed data types. To evaluate the clusters found; cluster distribution and Silhouette index were used for cluster quality, Cohen's Kappa Index for cluster stability and Cramer's V Coefficient for clinical relevance of clusters. Six distinct demographic subgroups were consistently found by three algorithms while model-based clustering solution were not considered for deeper analysis as highly imbalanced clusters were produced. The subgroups found have good quality clusters, moderate association with cancer types and high stability. The top three prevalent cancers associated with these subgroups were consistently identified using the three algorithms. Upon comparing the subgroups’ ages during diagnosis, we identify possible screening behaviours of specific subgroups, suggesting for early screening awareness programmes. This study demonstrates the use of cluster analysis in a cancer registry to identify demographic subgroups that could suggest potential areas to develop targeted and improved healthcare management strategies.

Разработка стратегий развития студенческих спортивных клубов на основе кластерного анализа

The development of student sports clubs is one of the key priorities in student sports, and in this direction an ambitious goal has been set to establish clubs in 100% of universities and colleges. To achieve this goal, a typical model of a student sports club was developed, however, at the current stage of development, not all clubs can apply it since they are characterized by a variety of organizational forms. To increase the effectiveness of the development of student sports clubs, it is necessary to improve approaches to strategic management, which would take into account differences of the level of clubs’ development. In this article, the author suggests recommendations for formulation of a student sports clubs development strategy based on cluster analysis. The results of the cluster analysis of student sports clubs revealed 5 clusters according to the level of development, and the belonging to a club in a certain cluster affects the determination of the potential stakeholders, the conduction of competitive analysis, the definition of strategic priorities, the development of functional strategies for student sports clubs. In this regard, the use of cluster analysis for the strategic management of student sports clubs is an effective tool, especially in the context of digitalization of management in student sports.

Assessment of the Impact of Commercial Banks’ Operating Activities on the Natural Environment by Use of Cluster Analysis

The aim of the paper is to identify groups of banks with similar environmental commitment, taking into account their direct environmental impact. The study, which employs the aggregation method, reveals that small banks with a relatively worse financial standing are characterised by the lowest level of disclosures within pro-ecological initiatives. At the same time, large international banks belong to clusters defined by the highest or the lowest disclosure rates. The above-mentioned phenomenon results from the dichotomy of integrating environmental policy into their strategies and business models. This study is the first comparative analysis of the extent to which all listed (and at the same time the biggest) banks operating in Poland have taken initiatives to reduce the negative environmental impact of their activities.