Cluster Analysis Approach Research Articles

Cytokine signature clusters as a tool to compare changes associated with tobacco product use in upper and lower airway samples.

Inhalation exposure to cigarette smoke and e-cigarette aerosol is known to alter the respiratory immune system, particularly cytokine signaling. In assessments of health impacts of tobacco product use, cytokines are often measured using a variety of sample types, from serum to airway mucosa. However, it is currently unclear whether and how well cytokine levels from different sample types and the airway locations they represent are correlated, making comparing studies that utilize differing sample types challenging. To address this challenge, we compared baseline cytokine signatures in upper and lower airways and systemic samples and evaluated how groups of coexpressed cytokines change with tobacco product use. Matched nasal lavage fluid (NLF), nasal epithelial lining fluid (NELF), sputum, and circulating serum samples were collected from 14 nonsmokers, 13 cigarette smokers, and 17 e-cigarette users and analyzed for levels of 22 cytokines. Individual cytokine signatures were first compared across each sample type, followed by identification of cytokine clusters within each sample type. Identified clusters were then evaluated for potential alterations following tobacco product use using eigenvector analyses. Individual cytokine signatures in the respiratory tract were significantly correlated (NLF, NELF, and sputum) compared with randomly permutated signatures, whereas serum was not significantly different from random permutations. Cytokine clusters that were similar across airway sample types were modified by tobacco product use, particularly e-cigarettes, indicating a degree of uniformity in terms of how cytokine host defense and immune cell recruitment responses cooperate in the upper and lower airways. Overall, cluster-based analyses were found to be especially useful in small cohort assessments, providing higher sensitivity than individual signatures to detect biologically meaningful differences between tobacco use groups. This novel cluster analysis approach revealed that eigencytokine patterns in noninvasive upper airway samples simulate cytokine patterns in lower airways.

Practical and theoretical considerations of the effects of ionic liquids on the separation properties of phenyl-based stationary phases in reversed-phase liquid chromatography

The use of ionic liquids (ILs) as mobile phase additives in reversed-phase liquid chromatography (RP-LC) can influence the dynamic coating of the stationary phases and even allow them to change their chromatographic properties. In this study, the interaction between thirteen different ILs added to 0.1% HCOOH (mobile phase) and the functional groups on phenyl and pentafluorophenyl stationary phases was investigated via RP-LC analysis and the fluorescence detection (FL) of four basic aromatic analytes (anthracycline antibiotics). In addition, an octadecyl column was included in the experiments to investigate potential differences in how the aromatic and alkyl stationary phases interacted with the tested ILs. Changes in behaviors of the studied analytes in the absence and presence of tested ILs were evaluated based on retention factor (k), peak area (A), number of theoretical plates (NA) and tailing factor (Tf). Furthermore, hierarchical cluster analysis (HCA) and k-means approach were used to better visualize found relationships between the chromatographic parameters of anthracyclines for each column. The obtained results indicated that π-π interactions occurring on the phenyl-based stationary phases resulted in more pronounced effects from the ILs compared to those occurring on the alkyl stationary phase. Different subclusters of the k, A, NA, and Tf parameters detected by the HCA confirmed that the addition of ILs leads to various changes in the chromatographic parameters of the studied analytes. These observations were additionally confirmed by k-means calculations. Overall, our findings show that selecting the appropriate IL for addition to the mobile phase can significantly reduce analysis time and improve peak shape and column performance when using aromatic stationary phases.

Revisiting Cluster Vulnerabilities towards Information and Communication Technologies in the Eastern Island of Indonesia Using Fuzzy C Means

Design/methodology/approach: In the present digital era, technology infrastructure plays an important role in the development of digital literacy in various sectors that can provide various important information on a large scale. Purpose: The use of information and communication technology (ICT) in Indonesia in the last five years has shown a massive development of ICT indicators. The population using the internet also experienced an increase during the period 2016–2020, as indicated by the increasing percentage of the population accessing the internet in 2016 from around 25.37 percent to 53.73 percent in 2020. This study led to a review of the level of ICT vulnerability in eastern Indonesia through a machine learning-based cluster analysis approach. Implications: Data were collected in this study from Badan Pusat Statistik (BPS) through SUSENAS to obtain an overview of the socioeconomic level and SAKERNAS to capture the employment side. This study uses 15 variables based on aspects of business vulnerability covering 174 districts/cities. Practical implications: Cluster analysis using Fuzzy C Means (FCM) was used to obtain a profile of ICT level vulnerability in eastern Indonesia by selecting the best model. The best model is obtained by selecting the validation value such as Silhouette Index, Partition Entropy, Partition Coefficient, and Modified Partition Coefficient. Social implication: For some areas with a very high level of vulnerability, special attention is needed for the central or local government to support the improvement of information technology through careful planning. Socio-economic and occupational aspects have been reflected in this very vulnerable cluster, and the impact of the increase in ICT will provide a positive value for community development. Originality/value: From the modelling results, the best cluster model is two clusters, which are categorized as high vulnerability and low vulnerability. For each cluster member who has a similarity or proximity to each other, there will be one cluster member.

Food Addiction in Eating Disorders: A Cluster Analysis Approach and Treatment Outcome.

Background: A first approach of a phenotypic characterization of food addiction (FA) found three clusters (dysfunctional, moderate and functional). Based on this previous classification, the aim of the present study is to explore treatment responses in the sample diagnosed with Eating Disorder(ED) of different FA profiles. Methods: The sample was composed of 157 ED patients with FA positive, 90 with bulimia nervosa (BN), 36 with binge eating disorder (BED), and 31 with other specified feeding or eating disorders (OSFED). Different clinical variables and outcome indicators were evaluated. Results: The clinical profile of the clusters present similar characteristics with the prior study, having the dysfunctional cluster the highest ED symptom levels, the worse psychopathology global state, and dysfunctional personality traits, while the functional one the lowest ED severity level, best psychological state, and more functional personality traits. The dysfunctional cluster was the one with lowest rates of full remission, the moderate one the higher rates of dropouts, and the functional one the highest of full remission. Conclusions: The results concerning treatment outcome were concordant with the severity of the FA clusters, being that the dysfunctional and moderate ones had worst treatment responses than the functional one.

Assessing Tumour Haemodynamic Heterogeneity and Response to Choline Kinase Inhibition Using Clustered Dynamic Contrast Enhanced MRI Parameters in Rodent Models of Glioblastoma.

Simple SummaryThis study was designed to monitor changes in DCE-MRI-based parameters in preclinical GBM models in response to choline kinase inhibition using a cluster analysis approach. In terms of therapeutic response in F98 rat GBMs, a sustained decrease in permeability and perfusion and increased necrosis was observed during treatment with JAS239 as compared to control animals. No significant differences in these parameters were found for the GL261 mice GBMs. The study demonstrates that region-based clustered pharmacokinetic parameters obtained using DCE-MRI can be used for detecting and assessing tumour haemodynamic heterogeneity, which may be useful in assessing therapeutic response.To investigate the utility of DCE-MRI derived pharmacokinetic parameters in evaluating tumour haemodynamic heterogeneity and treatment response in rodent models of glioblastoma, imaging was performed on intracranial F98 and GL261 glioblastoma bearing rodents. Clustering of the DCE-MRI-based parametric maps (using Tofts, extended Tofts, shutter speed, two-compartment, and the second generation shutter speed models) was performed using a hierarchical clustering algorithm, resulting in areas with poor fit (reflecting necrosis), low, medium, and high valued pixels representing parameters , , Kep, , and . There was a significant increase in the number of necrotic pixels with increasing tumour volume and a significant correlation between and tumour volume suggesting increased extracellular volume in larger tumours. In terms of therapeutic response in F98 rat GBMs, a sustained decrease in permeability and perfusion and a reduced cell density was observed during treatment with JAS239 based on , and as compared to control animals. No significant differences in these parameters were found for the GL261 tumour, indicating that this model may be less sensitive to JAS239 treatment regarding changes in vascular parameters. This study demonstrates that region-based clustered pharmacokinetic parameters derived from DCE-MRI may be useful in assessing tumour haemodynamic heterogeneity with the potential for assessing therapeutic response.

Middle School Students' Mathematical Resilience and Perceptions of Mathematics: A Cluster Analysis Approach

The primary purpose of the present study was to examine whether Turkish middle school students' perceptions of mathematics teachers, perceptions of mathematics learning, and their mathematical resilience can be used to divide students into clusters with similar profiles. The sample consisted of middle school students with ages ranging from 11 to 15. Two-step cluster analysis was used to create profile groups. As a result of cluster analysis, two distinct profiles were revealed. The results showed that clusters -formed in the present study differentiated according to students’ perceptions of mathematics teachers, learning mathematics, their mathematical resilience, and their mathematics achievement. Students in cluster 1 have higher mathematics achievement, value mathematics, perceive mathematics teachers as knowledgeable and supportive people, perceive learning mathematics as a fun process requiring effort. Students in cluster 2 have lower mathematics achievement. They perceive mathematics teachers as the source of anxiety and learning mathematics as a difficult process. They also have low belief levels that mathematics ability can be improved in comparison to cluster 1.

Exploring dietary quality characteristics related to milk yield of crossbred cows in Ethiopian smallholder farms: a clustering and multivariate analysis approach.

The study, conducted on 70 smallholder dairy farms in Northern Ethiopia, aimed to evaluate whether variation in milk yield (in early and mid-lactation) of multiparous Holstein-Friesian crossbred cows is related to diet composition and quality. At early stage (1-120days in milk (DIM)), a total of 70 dairy farms were used, while at mid-lactation (121-240 DIM), 54 dairy farms continued to be part of the study. K-means clustering was applied to group the cows based on energy-corrected milk yield (ECMY) into three milk production farm clusters (MPFC): Low MPFC (5.7-9.3 L/day), medium MPFC (9.4-12.8 L/day), and high MPFC (12.9-17.6 L/day). The dry matter intake (DMI) of cows during early lactation for high MPFC and low MPFC was 14.1 and 11.2kg/day, respectively. The dietary proportion of crop residues in diets offered to crossbred cows tended to be lower in the high MPFC during early as well as in mid-lactation. Cows from the high MPFC consumed diets with higher (rumen degradable) protein levels both in early and in mid-lactation, while dietary fiber fractions and in vitro dry matter digestibility (IVDMD) only differed in early lactation. Multiple regression models indicated that DMI (kg/day) in combination with either neutral detergent fiber, crude protein, or IVDMD (g/kg DM) explained about 25% of the variation in daily ECMY expressed relative to body weight (mL/kg). Hence, higher milk production is linked to both increased DMI and better quality of diets.

Identifying patterns in trunk/head/elbow changes of riders and non-riders: A cluster analysis approach

Correct rider oscillation and position are the basics for a good horseback riding performance. In this paper, we propose a framework for the automatic analysis of athletes behaviour based on cluster analysis. Two groups of athletes (riders vs non-riders) were assigned to a horseback riding simulator exercise. The participants exercised four different incremental horse oscillation frequencies. This paper studies the postural coordination, by computing the different discrete relative phases of head-horse, elbow-horse and trunk-horse oscillations. Two clustering algorithms are then applied to automatically identify the change of rider and non-rider behaviour in terms of postural coordination. The results showed that the postural coordination was influenced by the level of rider expertise. More diverse behaviour was observed for non-riders. At the opposite, riders produced lower postural displacements and deployed more efficient postural control. The postural coordination for both groups was also influenced by the oscillation frequencies.

Mobile Dating Applications and the Sexual Self: A Cluster Analysis of Users' Characteristics.

The online environment has had a profound sociocultural impact, and its implications pose new challenges to modern-day societies. The digital sexual and dating domains have dramatically affected sexual and romantic behavior and norms, and yet, no empirical studies have addressed the role of the sexual self-concept in driving sexual and romantic online behavior. The aim of this study is to identify reliable sexual self-configurations through a cluster analysis approach and determine whether these profiles are informative in relation to app use motives and sexual outcomes. For this purpose, a total of 244 subjects were recruited. Survey questions measured a set of demographic parameters, motives for app use, online sexual activities, attachment orientations, the sexual self-concept and sexual satisfaction. Five clusters were identified, including two with high levels of sexual drive (“Driven” and “Adventurous”), two with low levels of sexual drive (“Unassured” and “Naïve”) and one with an intermediate level of sexual drive (“Competent”). The clusters differed on gender, attachment styles, app use patterns and sexual characteristics. These findings provide insight into how the sexual self-concept relates to the interaction between individuals and the online sexual and dating scenes.

Year-Round Dive Characteristics of Male Beluga Whales From the Eastern Beaufort Sea Population Indicate Seasonal Shifts in Foraging Strategies

Dive behavior represents multiple ecological functions for marine mammals, but our understanding of dive characteristics is typically limited by the resolution or longevity of tagging studies. Knowledge on the time-depth structures of dives can provide insight into the behaviors represented by vertical movements; furthering our understanding of the ecological importance of habitats occupied, seasonal shifts in activity, and the energetic consequences of targeting prey at a given depth. Given our incomplete understanding of Eastern Beaufort Sea (EBS) beluga whale behavior over an annual cycle, we aimed to characterize dives made by belugas, with a focus on analyzing shifts in foraging strategies. Objectives were to (i) characterize and classify the range of beluga-specific dive types over an annual cycle, (ii) propose dive functions based on optimal foraging theory, physiology, and association with environmental variables, and (iii) identify whether belugas undergo seasonal shifts in the frequency of dives associated with variable foraging strategies. Satellite-linked time-depth-recorders (TDRs) were attached to 13 male belugas from the EBS population in 2018 and 2019, and depth data were collected in time series at a 75 s sampling interval. Tags collected data for between 13 and 357 days, including three tags which collected data across all months. A total of 90,211 dives were identified and characterized by twelve time and depth metrics and classified into eight dive types using a Gaussian mixed modeling and hierarchical clustering analysis approach. Dive structures identify various seasonal behaviors and indicate year-round foraging. Shallower and more frequent diving during winter in the Bering Sea indicate foraging may be energetically cheaper, but less rewarding than deeper diving during summer in the Beaufort Sea and Arctic Archipelago, which frequently exceeded the aerobic dive limit previously calculated for this population. Structure, frequency and association with environmental variables supports the use of other dives in recovery, transiting, and navigating through sea ice. The current study provides the first comprehensive description of the year-round dive structures of any beluga population, providing baseline information to allow improved characterization and to monitor how this population may respond to environmental change and increasing anthropogenic stressors.

Examining the attitude of Indian consumers toward retail loyalty programs: a cluster analysis approach

Examining the attitude of Indian consumers toward retail loyalty programs: a cluster analysis approach

Examining Attitude of Indian Consumers toward Retail Loyalty Programs: A Cluster Analysis Approach

Examining Attitude of Indian Consumers toward Retail Loyalty Programs: A Cluster Analysis Approach

Interactive Visual Cluster Analysis by Contrastive Dimensionality Reduction.

We propose a contrastive dimensionality reduction approach (CDR) for interactive visual cluster analysis. Although dimensionality reduction of high-dimensional data is widely used in visual cluster analysis in conjunction with scatterplots, there are several limitations on effective visual cluster analysis. First, it is non-trivial for an embedding to present clear visual cluster separation when keeping neighborhood structures. Second, as cluster analysis is a subjective task, user steering is required. However, it is also non-trivial to enable interactions in dimensionality reduction. To tackle these problems, we introduce contrastive learning into dimensionality reduction for high-quality embedding. We then redefine the gradient of the loss function to the negative pairs to enhance the visual cluster separation of embedding results. Based on the contrastive learning scheme, we employ link-based interactions to steer embeddings. After that, we implement a prototype visual interface that integrates the proposed algorithms and a set of visualizations. Quantitative experiments demonstrate that CDR outperforms existing techniques in terms of preserving correct neighborhood structures and improving visual cluster separation. The ablation experiment demonstrates the effectiveness of gradient redefinition. The user study verifies that CDR outperforms t-SNE and UMAP in the task of cluster identification. We also showcase two use cases on real-world datasets to present the effectiveness of link-based interactions.

The Bio-Geographical Regions Division of Global Terrestrial Animal by Multivariate Similarity Clustering Analysis Method

A novel multivariate similarity clustering analysis (MSCA) approach was used to estimate a biogeographical division scheme for the global terrestrial fauna and was compared against other widely used clustering algorithms. The faunal dataset included almost all terrestrial and freshwater fauna, a total of 4631 families, 141,814 genera, and 1,334,834 species. Our findings demonstrated that suitable results were only obtained with the MSCA method, which was associated with distinct hierarchies, reasonable structuring, and furthermore, conformed to biogeographical criteria. A total of seven kingdoms and 20 sub-kingdoms were identified. We discovered that the clustering results for the higher and lower animals did not differ significantly, leading us to consider that the analysis result is convincing as the first zoogeographical division scheme for global all terrestrial animals.

Feature Selection for Cluster Analysis in Spectroscopy

Cluster analysis in spectroscopy presents some unique challenges due to the specific data characteristics in spectroscopy, namely, high dimensionality and small sample size. In order to improve cluster analysis outcomes, feature selection can be used to remove redundant or irrelevant features and reduce the dimensionality. However, for cluster analysis, this must be done in an unsupervised manner without the benefit of data labels. This paper presents a novel feature selection approach for cluster analysis, utilizing clusterability metrics to remove features that least contribute to a dataset's tendency to cluster. Two versions are presented and evaluated: The Hopkins clusterability filter which utilizes the Hopkins test for spatial randomness and the Dip clusterability filter which utilizes the Dip test for unimodality. These new techniques, along with a range of existing filter and wrapper feature selection techniques were evaluated on eleven real-world spectroscopy datasets using internal and external clustering indices. Our newly proposed Hopkins clusterability filter performed the best of the six filter techniques evaluated. However, it was observed that results varied greatly for different techniques depending on the specifics of the dataset and the number of features selected, with significant instability observed for most techniques at low numbers of features. It was identified that the genetic algorithm wrapper technique avoided this instability, performed consistently across all datasets and resulted in better results on average than utilizing the all the features in the spectra.

CLUSTER ANALYSIS IDENTIFIES VARIABLES RELATED TO PROGNOSIS OF BREAST CANCER DISEASE

This work presents a cluster analysis approach aiming to determine distinct groups based on clinicopathological data from patients with breast cancer (BC). For this purpose, the clinical variables were considered: age at diagnosis, weight, height, lymph nodal invasion (LN), tumor-node-metastasis (TNM) staging and body mass index (BMI). Ward's hierarchical clustering algorithm was used to form specific groups. Based on this, BC patients were separated into four groups. The Kruskal-Wallis test was performed to assess the differences among the clusters. The intensity of the influence of variables on the prognosis of BC was also evaluated by calculating the Spearman's correlation. Positive correlations were obtained between weight and BMI, TNM and LN invasion in all analyzes. Negative correlations between BMI and height were obtained in some of the analyzes. Finally, a new correlation was obtained, based on this approach, between weight and TNM, demonstrating that the trophic-adipose status of BC patients can be directly related to disease staging.

Do COVID-19-Related Stress, Being Overweight, and Body Dissatisfaction Contribute to More Disordered Eating in Polish Women?-A Cluster Analysis Approach.

We hypothesized that women who are overweight, experiencing COVID-19-related stress, and with high body dissatisfaction would have significantly greater disordered eating than those of healthy weight, without stress, and with low body dissatisfaction. Participants (N = 1354 women; Mage= 31.89 years, SD = 11.14) filled in the Contour Drawing Rating Scale, the Emotional Overeating Questionnaire, the Eating Motivation Survey, the Mindful Eating Questionnaire, and a COVID-19-related stress measure and sociodemographic survey. The cluster analysis technique revealed four distinct clusters: (a) Cluster 1 (N = 314): healthy body weight, no COVID-related stress, and low body dissatisfaction (M = 1.19); (b) Cluster 2 (N = 131): overweight, no COVID-related stress, and high body dissatisfaction (M = 2.41); (c) Cluster 3 (N = 597): healthy body weight, COVID-related stress, and low body dissatisfaction (M = 1.27); (d) Cluster 4 (N = 312): overweight, COVID-related stress, and high body dissatisfaction (M = 2.84). Generally, our outcomes partially support our hypothesis, as higher levels of some types of disordered eating were observed in women who were overweight with COVID-related stress and high body dissatisfaction (Cluster 4) as compared with women with healthy body weight, no COVID-related stress, and with low levels of body dissatisfaction (Cluster 1). Our results indicate that both body weight status, as well as COVID-19-related stress and body dissatisfaction, may contribute to the intensity of disordered eating. During future epidemic-related quarantines, this may be an argument in favor of organizing support regarding emotional functioning, body image, and eating behaviors, particularly for the most vulnerable groups—including overweight and obese women.

High School Students’ Accelerometer-Measured Physical Activity and Sedentary Behavior by Motivational Profiles Toward Physical Activity

ABSTRACT Purpose: Based on a Self-Determination Theory with a person-centered approach, the purposes were: (1) to identify motivational profiles toward physical activity (PA) among high school students, and (2) to compare students’ accelerometer-measured PA and sedentary behavior (SB) levels between the motivational profiles toward PA. Methods: The present study followed a cross-sectional design. A total of 394 high school students (173 females; final sample = 360 and 198 for the purposes 1 and 2, respectively) aged 12–16 years completed the study. Participants’ PA and SB levels were objectively measured by ActiGraph GT3X+ accelerometers (right hip) during seven consecutive days. Later, motivation toward PA was measured by the Behavioral Regulation in Exercise Questionnaire. Results: Based on a two-stage cluster analysis approach, the results identified three clusters: (1) “Self-determined motivational profile toward PA” (65.3%); (2) “Non-self-determined motivational profile toward PA” (23.9%), and (3) “Controlled motivational profile toward PA” (10.8%). The results of the chi-square test showed that a statistically significantly higher percentage of males were represented in cluster 1 (p < .001). The results of the one-way analysis of variance showed that students from cluster 1 had statistically significantly lower levels of SB, and higher levels of light PA, moderate-to-vigorous PA and total PA than those from cluster 2 (p < .05; |d| = 0.40–0.61). Conclusions: In line with the Self-Determination Theory, the students within the self-determined motivational profile had lower SB and higher PA. This preliminary evidence could inform and guide future policy regarding students’ PA promotion.

Comparing the utilities consumptions in Dubai per category and community: MANOVA and cluster analysis approaches

UAE is marked by the increasing demand for water and electricity due to demographic, environmental and economic factors, coupled with the dependence on water desalination process, which is costly, consumes a lot of energy and is non-environmentally friendly. Like most of the authorities in UAE, Dubai Electricity and Water Authority is facing the challenges of balancing supply with demand and responding to consumer requirements, from one side, and addressing the continuously increasing consumption and slowing it down from another side. Therefore, policy makers can benefit from statistical data analysis in order to make informed decisions. This study aims to equip decision makers with useful tools and analysis to address some of their short- and long-term objectives related to production and consumption. The current study focused on three main objectives: (i) analysing the production of the desalination plants in Dubai, (ii) comparing the consumptions of water and electricity based on the four categories residential, commercial, industrial and others, and (iii) segmenting the various communities in Dubai depending on their consumption behavior. The data used for this study is collected from the open government data and SAS Programming is adopted for data analysis. The results of the analysis revealed that the desalinated water production follows an upward trend, yet still not in line with the consumption growth. Furthermore, there are significant differences between the four categories for both water and electricity consumptions. The highest levels of consumptions are associated with the residential and commercial categories. Finally, the cluster analysis technique revealed fifteen clusters of communities depending on the consumption levels.

An overview of the academic level among EU countries. A cluster analysis approach

Abstract The modernization of higher education is a complex process and occupies a priority place on the agenda of concerns both at European level and at the level of each member state of the European Union (EU). In this paper we present the results of the analysis of the main aspects of post-integration modernization of higher education in Romania, compared to the evolution in other Member States in the same period, from the perspective of five indicators: expenditure per student, percentage of GDP allocated to higher education, graduates per thousand inhabitants and professor per student’s ratio. In order to achieve the proposed objective, we have carried out a research and a cluster analysis of data from secondary sources of available information, respectively reports of the European Commission and various statistical reports that will be mentioned throughout the article.