Application Of Cluster Analysis Research Articles

Neuropsychological Assessments of Patients With Acquired Brain Injury: A Cluster Analysis Approach to Address Heterogeneity in Web-Based Cognitive Rehabilitation.

We aimed to (1) apply cluster analysis techniques to mixed-type data (numerical and categorical) from baseline neuropsychological standard and widely used assessments of patients with acquired brain injury (ABI) (2) apply state-of-the-art cluster validity indexes (CVI) to assess their internal validity (3) study their external validity considering relevant aspects of ABI rehabilitation such as functional independence measure (FIM) in activities of daily life assessment (4) characterize the identified profiles by using demographic and clinically relevant variables and (5) extend the external validation of the obtained clusters to all cognitive rehabilitation tasks executed by the participants in a web-based cognitive rehabilitation platform (GNPT). We analyzed 1,107 patients with ABI, 58.1% traumatic brain injury (TBI), 21.8% stroke and 20.1% other ABIs (e.g., brain tumors, anoxia, infections) that have undergone inpatient GNPT cognitive rehabilitation from September 2008 to January 2021. We applied the k-prototypes algorithm from the clustMixType R package. We optimized seven CVIs and applied bootstrap resampling to assess clusters stability (fpc R package). Clusters' post hoc comparisons were performed using the Wilcoxon ranked test, paired t-test or Chi-square test when appropriate. We identified a three-clusters optimal solution, with strong stability (>0.85) and structure (e.g., Silhouette > 0.60, Gamma > 0.83), characterized by distinctive level of performance in all neuropsychological tests, demographics, FIM, response to GNPT tasks and tests normative data (e.g., the 3 min cut-off in Trail Making Test-B). Cluster 1 was characterized by severe cognitive impairment (N = 254, 22.9%) the mean age was 47 years, 68.5% patients with TBI and 22% with stroke. Cluster 2 was characterized by mild cognitive impairment (N = 376, 33.9%) mean age 54 years, 53.5% patients with stroke and 27% other ABI. Cluster 3, moderate cognitive impairment (N = 477, 43.2%) mean age 33 years, 83% patients with TBI and 14% other ABI. Post hoc analysis on cognitive FIM supported a significant higher performance of Cluster 2 vs. Cluster 3 (p < 0.001), Cluster 2 vs. Cluster 1 (p < 0.001) and Cluster 3 vs. Cluster 1 (p < 0.001). All patients executed 286,798 GNPT tasks, with performance significantly higher in Cluster 2 and 3 vs. Cluster 1 (p < 0.001).

Evaluation of genetic diversity of subdivided genealogical groups in Lithuanian Trakehner horse population using immunogenetic tools

The objective of this study was to evaluate the inter-breed genetic diversity of Lithuanian Trakehner horses. The current population in Lithuania numbers 1 039 Trakehner horses. The study included the immunogenetic data analysis of 316 blood samples. Pedigree was traced back to 7–11 generations. The horses were assigned to genealogical clusters originating from East Prussian Trakehner Stud, Thoroughbred and Arab sires. The first two clusters were subdivided into nine-horse groups according to the most important ancestors. The evaluation of genetic diversity was based on six blood groups and five protein polymorphism systems. The differences between genealogical groups were confirmed by principal component analysis and applied cluster analysis. According to the allelic frequency, the Trakehner population changed very slightly through a 30-year period, and only one rare (0.013) allele Tf&lt;sup&gt;RR&lt;/sup&gt; (frequent in other Trakehner populations), has been newly found. The analysis of genetic polymorphism systems based on a very low rate of alleles Q&lt;sup&gt;abc&lt;/sup&gt; (0.093) and D&lt;sup&gt;dkl&lt;/sup&gt; (0.046), which are typical of other populations, indicated the exclusivity of the Trakehner population in Lithuania. The average expected heterozygosity by blood protein polymorphism and blood groups were 0.409 and 0.441, respectively. The genetic clustering diverged by observed heterozygosity of blood groups and by pedigree data in Pilger through Egoist, Bay Ronald through Dark Ronald and Dampfross through Hyperion subgroups. The determined distinctiveness of subdivided Trakehner horse groups suggests reconsidering the breeding strategies and conservation programme of Trakehner horses. Breeding and selection of subdivided sire lines could be among the appropriate solutions for the maintenance and extension of the genealogical structure.

Defective democracy and the political budget cycle

Using a panel of democracies from 1987 to 2016, this study examines the effect of a defective democracy on the political budget cycle. To separate defective democracies from embedded democracies, we apply cluster analysis to the central elements of democracy. We find that the political budget cycle is significant only in defective democracies. Our finding indicates that the electoral cycle can persist even though democracies become old and economies become advanced. We also find that, of the central elements of democracy, weak rule of law and an ineffective government mainly explain the political budget cycle in defective democracies.

Variability of Morphological Traits of European Beech (Fagus sylvatica L.) Seedlings in Serbia

The results of interpopulation variability of morphometric parameters of European beech (Fagus sylvatica L.) seedlings originating from eight populations from the part of natural distribution area in Serbia are presented in this paper. The studied populations of Dubašnica, Jastrebac, Boranja, Fruška Gora, Mali Pek, Goč, Beljanica, and Javor have various ecological and vegetational characteristics. The results of this paper refer to root collar diameter and height of seedlings at the age of 1+0 and 2+0. Analysis of variance showed a statistically significant difference between the populations in terms of the studied morphological traits of seedlings (p&amp;lt;0.01; α=0.05). Geographical differentiation of the studied populations has not been determined by applying cluster analysis, but the populations are grouped randomly and they indicate the ecotypic nature of beech genetic variation. The results of this research may serve in beech breeding and the available gene pool conservation. Based on the obtained results it can be recommended that in terms of transfer and use of the beech forest reproductive material greater attention should be paid to the ecological conditions of the parent stands and habitats where afforestation has been performed.

Influence of the modes of climate variability in the Tropical Pacific and Atlantic on accumulated rainfall and reservoir water volumes in the Northeast Brazil

AbstractThe volumes of artificial reservoirs as well as the accumulated rainfall over the Northeast region of Brazil (NEB) are sensible to the modes of climate variability. Thus, the objective of this study is to identify the relationship between the spatial patterns of the modes of climate variability in the Tropical Atlantic and Pacific Oceans, and the homogeneous rainfall subregions affecting the recharge of the water volumes of reservoirs in the NEB during the period from 1986 to 2018. For this end, we used monthly rainfall; monthly water volume data for artificial reservoirs; sea surface temperature anomalies in the Tropical Pacific and Tropical Atlantic Ocean. The first step was to apply cluster analysis to subdivide the NEB according to the monthly rainfall climatology. After, the technique of synthetic maximum rainfall series was applied. Then, the response time between rainfall climatology and reservoir water volume climatology was identified by means of cross‐correlation. Finally, the partial least squares regression was applied to infer the relationship between synthetic maximum rainfall series in each subregion and the modes of climate variability. Results showed that five homogeneous rainfall subregions with three distinct wet seasons. The synthetic time series of rainfall in the subregions were positively correlated with the climatology of the reservoir volumes with lags of up to 5 months after the beginning of the series. Both rainfall and the water volume of artificial reservoirs were sensible to the modes of climate variability of the studied oceans.

Developing Insights for Possible and Probable Acute Concussions Using Cluster Analysis.

Few studies have analyzed the Sport Concussion Assessment Tool's (SCAT) utility among athletes whose concussion assessment is challenging. Using a previously published algorithm, we identified possible and probable concussions at <6 h (n = 393 males, n = 265 females) and 24-48 h (n = 323 males, n = 236 females) post-injury within collegiate student-athletes and cadets from the Concussion Assessment, Research, and Education (CARE) Consortium. We applied cluster analysis to characterize performance on the Standard Assessment of Concussion (SAC), Balance Error Scoring System (BESS), and the SCAT symptom checklist for these athletes. Among the cluster sets that best separated acute concussions and normal performances, total symptom number raw score and change and post-traumatic migraine raw score and change score were the most frequent clustering variables across males and females at <6 h and 24-48 h. Similarly, total symptom number raw score and change score and post-traumatic migraine raw score and change score were most significantly different between clusters for males and females at <6 h and 24-48 h. Our results suggest that clinicians should focus on total symptom number, post-traumatic migraine symptoms, and cognitive-fatigue symptoms when assessing possible and probable concussions, followed by the SAC and BESS scores.

A brachiopod fauna from latest Permian to Induan of northern Guizhou, South China and its evolutionary pattern

A new brachiopod fauna is reported from the Bozhou section spanning the latest Permian and Early Induan in Zunyi, northern Guizhou, South China. The brachiopods were collected from the Changhsingian Changhsing Formation and the Lower Induan Yelang Formation. The Changhsing Formation is composed of limestone and bioclastic limestone. The lower part of the Yelang Formation is composed of calcareous mudstone and argillaceous limestone intercalated with clay, and the upper part of the Yelang Formation comprises limestone intercalated with calcareous mudstone. After applying cluster analysis on the occurrence data of brachiopod species, three communities are recognized. FAS community mainly includes Fusichonetes, Acosarina, Spinomarginifera, and Araxathyris individuals. LP community mainly contains several Permian holdovers including Paracrurithyris, Orbicoelia, and Araxathyris, and disaster taxa like Lingularia. AP community is composed of Abrekia?, Paracrurithyris, and Prelissorhynchia. Shift of brachiopods from FAS community to LP community represents the extinction stage. LP community represents the survival stage. AP community is in the survival‐initial recovery stage. In the extinction stage, global brachiopod faunas showed similar patterns in that most brachiopod species and genera disappeared. In the survival stage, Lingularia quickly occupied the vacant ecospace and dominated the brachiopod community in the shallow‐water environment with soft substrate. In the survival‐initial recovery stage, Mesozoic‐type brachiopods appeared and were sporadically found in various environments.

IDENTIFICATION OF AREAS OF CORONAVIRUS COVID-19 INCIDENCE SPREADING BASED ON CLUSTER ANALYSIS METHOD

Subject: the use of the mathematical apparatus of neural networks for the scientific substantiation of anti-epidemic measures in order to reduce the incidence of diseases when making effective management decisions. Purpose: to apply cluster analysis, based on a neural network, to solve the problem of identifying areas of incidence. Tasks: to analyze methods of data analysis to solve the clustering problem; to develop a neural network method for clustering the territory of Ukraine according to the nature of the epidemic process COVID-19; on the basis of the developed method, to implement a data analysis software product to identify the areas of incidence of the disease using the example of the coronavirus COVID-19. Methods: models and methods of data analysis, models and methods of systems theory (based on the information approach), machine learning methods, in particular the Adaptive Boosting method (based on the gradient descent method), methods for training neural networks. Results: we used the data of the Center for Public Health of the Ministry of Health of Ukraine distributed over the regions of Ukraine on the incidence of COVID-19, the number of laboratory examined persons, the number of laboratory tests performed by PCR and ELISA methods, the number of laboratory tests of IgA, IgM, IgG; the model used data from March 2020 to December 2020, the modeling did not take into account data from the temporarily occupied territories of Ukraine; for cluster analysis, a neural network of 60 input neurons, 100 hidden neurons with an activation Fermi function and 4 output neurons was built; for the software implementation of the model, the programming language Python was used. Conclusions: analysis of methods for constructing neural networks; analysis of training methods for neural networks, including the use of the gradient descent method for the Adaptive Boosting method; all theoretical information described in this work was used to implement a software product for processing test data for COVID-19 in Ukraine; the division of the regions of Ukraine into zones of infection with the COVID-19 virus was carried out and a map of this division was presented.

Multiple Evidence for Climate Patterns Influencing Ecosystem Productivity across Spatial Gradients in the Venice Lagoon

Effects of climatic changes in transitional ecosystems are often not linear, with some areas likely experiencing faster or more intense responses, which something important to consider in the perspective of climate forecasting. In this study of the Venice lagoon, time series of the past decade were used, and primary productivity was estimated from hourly oxygen data using a published model. Temporal and spatial patterns of water temperature, salinity and productivity time series were identified by applying clustering analysis. Phytoplankton and nutrient data from long-term surveys were correlated to primary productivity model outputs. pmax, the maximum oxygen production rate in a given day, was found to positively correlate with plankton variables measured in surveys. Clustering analysis showed the occurrence of summer heatwaves in 2008, 2013, 2015 and 2018 and three warm prolonged summers (2012, 2017, 2019) coincided with lower summer pmax values. Spatial effects in terms of temperature were found with segregation between confined and open areas, although the patterns varied from year to year. Production and respiration differences showed that the lagoon, despite seasonality, was overall heterotrophic, with internal water bodies having greater values of heterotrophy. Warm, dry years with high salinity had lower degrees of summer autotrophy.

Successful retailer strategies in price comparison platforms

The choice of an appropriate e‐commerce strategy for the listing in price comparison platforms (eBay, Amazon, and price search engines) is crucial for the survival of online stores in B2C e‐commerce business. We use a comprehensive dataset from the Austrian price search engine geizhals.at to identify successful e‐commerce strategies with regard to these listing decisions. An e‐commerce strategy is a set of choices including the listing decision, availability decision, and decisions on a price path and shipping cost. We apply cluster analysis to identify the different strategies that have been used by online retailers. Using various success measures such as revenue, clicks, market share, and the survival of firms, as dependent variables in our regression analyses, we present causal evidence on the effectiveness of different e‐commerce strategies.

Recovery of Forest Vegetation in a Burnt Area in the Republic of Korea: A Perspective Based on Sentinel-2 Data

Forest fires are severe disasters that cause significant damage in the Republic of Korea and the entire world, and an effort is being made to prevent forest fires internationally. The Republic of Korea budgets 3.38 million USD every year to prevent forest fires. However, an average of 430 wildfires occur nationwide annually. Thirty-eight percent of the forest fire budget is used for forest restoration. Restoring afforestation in the affected areas is a top priority. This study aimed to estimate the degree of vegetative regeneration using the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Soil-Adjustment Vegetation Index (SAVI), and Normalized Burn Ratio (NBR). Although many studies have used NBR with NDVI to extract plant regeneration regions, they suffer from atmospheric effects and soil brightness. Thus, this study utilizes NBR with NDVI, EVI, and SAVI to accurately select areas for targeted forest restoration. Furthermore, this study applies clustering analysis to extract the spatial boundary of vegetative regenerative regions. The proposed method suggests a pixel range of vegetation indices. These ranges can be used as an indicator, such as the NBR’s Fire Severity Level, which reflects the mountain’s local characteristics, meaning that it can be useful after forest fires. Using the three vegetation indices can extract more accurate vegetation areas than using NBR with NDVI and can help determine a forest restoration target area.

Groundwater Quality Assessment and Prediction of Spatial Variations in the Area of the Danube River Basin (Serbia)

Monitoring and forecasting of chemical and physicochemical parameters of groundwater is an important factor in quality control and water management. In order to optimize these processes, the initial purpose of this paper was to identify sources of pollution and predict spatial changes in groundwater quality. Patterns of spatial changes in groundwater quality in the Danube river basin (Serbia) have been identified using multivariate statistical techniques. The results of the applied cluster analysis are the indicators of the existance of two spatial clusters. The principal component/factor analysis (PCA/FA) has shown that beside natural, anthropogenic factor has an influence in spatial grouping. Discriminant analysis (DA) was applied in order to identify discriminant groundwater quality parameters. DA reduced the number of data by extracting two parameters (iron and arsenic). The spatial distribution of identified dominant factors and discriminant parameters were graphically represented using GIS. Finally, the artificial neural network technique was used to test the ability to predict spatial changes in the values of discriminant parameters, and the reliability of this technique to predict the spatial variations of the two extracted variables has been proven.

Green supply chain management strategy and financial performance in the shipping industry

ABSTRACT This study empirically examines the relationship between green supply chain management (GSCM) strategy and financial performance for shipping firms. We collect global data from 289 shipping firms and apply cluster analysis and analysis of variance (ANOVA) to test the research hypotheses. We cluster shipping firms according to their level of GSCM strategy adoption, which ranges from reactive to proactive. We find that shipping firms with a proactive GSCM strategy perform better financially. The findings of this study can contribute to shipping firms’ environmental management strategy decisions and help tackle environmental problems while improving their financial outcome. This paper provides a discussion of managerial and theoretical implications and recommends future research directions.

Air Combat Assignment Problem Based on Bayesian Optimization Algorithm

In order to adapt to the changing battlefield situation and improve the combat effectiveness of air combat, the problem of air battle allocation based on Bayesian optimization algorithm (BOA) is studied. First, we discuss the number of fighters on both sides, and apply cluster analysis to divide our fighter into the same number of groups as the enemy. On this basis, we sort each of our fighters’ different advantages to the enemy fighters, and obtain a series of target allocation schemes for enemy attacks by first in first serviced criteria. Finally, the maximum advantage function is used as the target, and the BOA is used to optimize the model. The simulation results show that the established model has certain decision-making ability, and the BOA can converge to the global optimal solution at a faster speed, which can effectively solve the air combat task assignment problem.

Inclusive practice and quality of education and care in the Dutch hybrid early childhood education and care system

The present study examined how in the context of the hybrid, privatized and marketized Dutch early education and care system (ECEC), childcare organizations respond to the public task of supporting inclusion and equity in an increasingly diverse society. Applying cluster analysis on the organizational characteristics of a nationally representative sample of 117 centers providing education and care for 0- to 4-year-old children, three types of organizations were identified that differed strongly on cultural inclusion and observed quality in the classroom. Socially engaged (for-profit and not-for-profit) professional organizations served proportionally more children from low-SES and immigrant families, provided higher quality to these children, and were culturally more inclusive than both market-orientated and traditional professional-bureaucratic organizations. The findings are discussed with regard to the question how hybrid ECEC systems can be governed to optimally serve the public goals of inclusion and equity.

Does Ranked-Choice Voting Reduce Racial Polarization? A Clustering Approach to Ranked Ballot Data

Political scientists have long examined if ranked-choice voting (RCV) yields moderation in racial and ethnic conflict in the electoral arena. We advance the literature by introducing novel empirical and theoretical perspectives to analyze individual ranked ballots in RCV to answer this long-debated question. By applying cluster analysis and ecological inference to more than 5.5 million ranked preferences and 13,635 precinct-level data from Bay Area mayoral elections from 1990 to 2020, we quantify the degree of racially polarized voting in a manner that is comparable across electoral systems. We show that switching from first-past-the-post to RCV election does not seem to reduce the level of racially polarized voting for the overall electorate and for all pairs of racial and ethnic groups examined. We also examine the impact of strategic ballot concentration known as plumping on the analysis of rank ballot data, shedding new light on the research of RCV and political behavior.

Brexit spillovers through international trade and foreign investment: Empirical evidence from EU-27 and the UK

This study examines the Brexit spillovers upon the European Union Member States (MS) (EU-27) and the UK through two fundamental freedoms of regional integration: goods and services (international trade), and capital (foreign investment, FDI). We have applied cluster analysis and structural equation modelling on a strongly balanced panel of EU-27 and the UK. Both techniques explore two scenarios that focus on the performances achieved by the EU-MS in terms of GDP per capita and GDP growth, under the impact of trade and FDI, before and after the Brexit (1995-2019 and 2020-2025 periods). Our results show that the UK?s economy will be affected both related to GDP growth and GDP per capita levels, particularly on the short-run. The EU-27 impact largely differs across countries and types of international activities, being decisively influenced through the FDI relations. Overall, the spillovers induced by international flows are positive, but significantly diminished after the Brexit.

Recognition Methods for Coal and Coal Gangue Based on Deep Learning

Recognizing coal and coal gangue is an important part of the coal industry and is mainly conducted via human sorting at present. Consequently, considerable manpower is needed, which adds a burden to enterprises and results in low efficiency. As an important branch of artificial intelligence, deep learning has been widely applied in many fields, especially in machine vision and voice recognition, its performance is greatly improved compared with the performances of traditional learning methods, and it also has good a transfer learning ability. This paper proposed an improved YOLOv4 algorithm as a classic deep learning method for the intelligent and highly accurate recognition of coal and coal gangue. Compared to other algorithms, YOLOv4 has a better anchor value by applying cluster analysis to different data sets, a good anti-interference ability due to using the Laplacian operator and Gaussian filter to reduce the impacts from mine dust and shock and acquires richer detailed information by increasing the number of layers of the feature pyramid. The experimental results show that compared with the other four algorithms of YOLOv4, YOLOv3, SSD and Faster-RCNN, the improved YOLOv4 proposed in this paper exhibits better detection accuracy, a better detection speed and robust performance.

Cluster Analysis for IR and NIR Spectroscopy: Current Practices to Future Perspectives

Supervised machine learning techniques have become well established in the study of spectroscopy data. However, the unsupervised learning technique of cluster analysis hasn’t reached the same level maturity in chemometric analysis. This paper surveys recent studies which apply cluster analysis to NIR and IR spectroscopy data. In addition, we summarize the current practices in cluster analysis of spectroscopy and contrast these with cluster analysis literature from the machine learning and pattern recognition domain. This includes practices in data pre-processing, feature extraction, clustering distance metrics, clustering algorithms and validation techniques. Special consideration is given to the specific characteristics of IR and NIR spectroscopy data which typically includes high dimensionality and relatively low sample size. The findings highlighted a lack of quantitative analysis and evaluation in current practices for cluster analysis of IR and NIR spectroscopy data. With this in mind, we propose an analysis model or workflow with techniques specifically suited for cluster analysis of IR and NIR spectroscopy data along with a pragmatic application strategy.

Integrated Community Onsite Sanitation System for Close Loop Faecal Management

The achievement of safe sanitation access is about 7.42 % in 2018, that indicates high risk of faecal contamination from unmanaged onsite sanitation system and has not been fully supported by faecal sludge service. This paper described the study of characteristic of onsite sanitation and faecal sludge for review potential of integrating of sludge handling with improved communal wastewater treatment. Faecal sludge quality were observed from pit latrines and communal septic tank, then analysis method apply cluster analysis and through close loop approach consider treatment and disposal for resource recovery. The pilot model of combined treatment of wastewater and faecal sludge was developed by anaerobic digester with a maximum 5% sludge loading, then further treatment by biofilter system and sludge drying bed. The quality of treated water can meet COD and nitrogen as effluent standard, biosolid reached a stabilization index of 40 - 60%, water content of 50-60% and less Coliform bacteria content. The characteristics of faecal sludge strength to be handled will be influenced bytechnical aspect or environment. The integrated system could be applied to manage wastewater and faecal sludge and its by products, mainly for rural area has not been catered by faecal sludge service.