Applying K-means Clustering Research Articles

Improving subsurface structural interpretation in complex geological settings through geophysical imaging and machine learning

This study employs seismic refraction tomography (SRT) and electrical resistivity tomography (ERT) to assess subsurface geological conditions along the proposed Porsgrunn Highway in Norway. The primary objective is to analyze SRT and ERT tomograms to identify subsurface geological structures. However, interpreting tomograms is often limited by smoothed boundaries and reduced resolution. To address these challenges, we apply k-means clustering, a machine learning technique that groups data based on similarities in physical properties, to post-process the geophysical tomograms. This study pioneers the use of k-means clustering to interpret tomograms from SRT and ERT data in complex geological settings. We first evaluate the effectiveness of clustering techniques using numerical modeling for two geological scenarios: a horizontally layered case and a layered case with undulation and a fault structure. Utilizing automated methods (Elbow and Silhouette), we objectively determine the optimal number of clusters for each geophysical tomogram. Subsequently, we compare the performance of the k-means clustering algorithm with subjective expert interpretations and the Laplacian edge detection method. Borehole data validate the clustering results and confirm the effectiveness of optimal cluster selection techniques. The findings of this study demonstrate that k-means clustering significantly enhances the detection of geological structures by establishing clearer boundaries and minimizing noise interference, enabling more accurate fault zone delineation. Compared to traditional edge detection and subjective interpretation methods, k-means clustering offers a systematic and objective approach that improves consistency and reliability across diverse geological settings. Moreover, its automated classification of geophysical data into meaningful clusters enables efficient analysis of large datasets. This study underscores the value of integrating machine learning techniques with geophysical methods such as SRT and ERT to improve interpretability and accurately identify subsurface geological structures, particularly in fault zone identification.

Identification of Global Extended Pseudo Invariant Calibration Sites (EPICS) and Their Validation Using Radiometric Calibration Network (RadCalNet)

This study introduces a global land cover clustering using an unsupervised algorithm, incorporating the novel step of filtering data to retain only temporally stable pixels before applying K-means clustering. Unlike previous approaches that did not assess the pixel-level temporal stability, this method provides more reliable clustering results. The K-means identified 160 distinct clusters, with Cluster 13 Global Temporally Stable (Cluster 13-GTS) showing significant improvements in temporal stability. Compared to Cluster 13 Global (Cluster 13-G) from earlier research, Cluster 13-GTS reduced the coefficient of variation by up to 1% and increased the number of calibration locations from 23 to over 50. This study also validated these clusters using TOA reflectance from ground-truth measurements collected at the Radiometric Calibration Network (RadCalNet) Gobabeb (RCN-GONA) site, incorporating data from Landsat 8, Landsat 9, Sentinel-2A, and Sentinel-2B. The GONA Extended Pseudo Invariant Calibration Sites (EPICS) GONA-EPICS cluster used for the validation provided statistically comparable mean TOA reflectance to RCN-GONA, with a reduced chi-square test indicating minimal differences within the cluster’s uncertainty range. Notably, the difference in reflectance between RCN-GONA and GONA-EPICS was less than 0.023 units across all the bands. Although GONA-EPICS exhibited slightly higher uncertainty (6.4% to 10.3%) compared to RCN-GONA site (<5%), it offered advantages such as 80 potential calibration points per Landsat cycle and reduced temporal instability, and it provided alternatives to reduce the reliance on single sites like traditional PICS or RCN-GONA, making it a valuable tool for calibration efforts. These findings highlight the potential of the newly developed EPICS for radiometric calibration and stability monitoring of optical satellite sensors. Distributed across diverse regions, these global targets increase the number of calibration points available for any sensor in any orbital cycle, reducing the reliance on traditional PICS and offering more robust targets for radiometric calibration efforts.

Optimizing large-scale CO2 pipeline networks using a geospatial splitting approach

CO2 transport infrastructure is the backbone of carbon capture and storage (CCS) technology for the mitigation of carbon emissions and project deployment viability. In conventional large-scale CO2 pipeline network designs, the storage sites are generally assumed as the centroids of the major geologic basins, however, this approach might provide suboptimal solutions since the large extension of some storage formations significantly increases the length of the CO2 transportation networks. To address this situation and obtain optimal pipeline routes, we present a novel geospatial splitting framework that partitions large basins into multiple sub-sinks. In our approach, we used a large number of reservoir models varying petrophysical properties and CO2 injection rates to compute pressure plumes through numerical simulations, leading to the calculation of the number of subregions for each basin as a function of the extension of pressure interference areas and boundaries. Finally, we applied K-means clustering and Voronoi polygon algorithms to partition large basins into subregions and obtain their sink coordinates. To demonstrate the capability of the developed workflow, we investigated two CO2 pipeline network modeling case studies using our splitting approach: one regional case study focusing on the Intermountain West (I-West) region and one nationwide case study covering the lower 48 states in the U.S. In both case studies, we compared the optimal pipeline routes using the original and new storage locations and examined the major differences. The use of the developed geospatial approach resulted in both cases in a shortening of the total pipeline network length by 13% and 10%, compared to the pipeline modeling with the original basins, leading to cost reductions of 25% and 17%, respectively, demonstrating that the location of point sinks has a critical impact on the length and expenses of pipelines to efficiently transport CO2 to distant storage sites. Therefore, the workflow presented here contributes to the proper and realistic modeling of case studies that support decision-making in CCS deployment.

Typology of esports crowdfunding investors

PurposeOnline fundraising has become a rapidly growing alternative business financing method for esports. Despite this, a lack of attention has been given to its fans' motivations for participating in crowdfunding (CF) projects or behavioral intentions for spectatorship. To address this, the present study (1) categorizes esport CF investors' motivations based on the big five personality traits and (2) compares whether motivations and behavioral intentions for esports spectatorships vary depending on each group.Design/methodology/approachWe applied K-means clustering analysis to classify investors. ANOVA was used to verify whether there were differences in CF motivations and behavioral intentions.FindingsThe study identified six distinct investor segments. Overall, esports CF investors showed high agreeableness and conscientiousness with low neuroticism. Fanatic fans were the most motivated and had the highest spectatorship intentions, driven by strong openness. In contrast, recognition seekers and charity donors exhibited the lowest viewing intentions. Entertainment seekers were highly motivated by excitement and aesthetics, linking CF participation to a strong desire to watch events. Despite higher neuroticism, reward hunters also showed significant viewing intentions, driven by drama and aesthetics. These findings highlight the diverse motivations and their influence on esports spectatorship.Originality/valueThis study investigates the motivations and behavioral intentions of CF investors and spectatorship, which has substantive theoretical and managerial implications for the business of esports.

Clustering Of Sapodilla Fruit (Manilkara Zapota) Maturity Levels Based On Color Using K-Means Clustering Method

The study aimed to evaluate the accuracy of determining sapodilla fruit maturity using the K-Means Clustering algorithm, a method that partitions data into clusters of similar characteristics; by applying K-Means Clustering on data samples obtained from Kaggle, and using Matlab for pixel value calculation, the algorithm effectively classified 150 sapodilla fruit samples into two clusters—75 mature and 75 raw—with an accuracy of 92.2% for mature sapodilla and 94.5% for raw sapodilla, demonstrating that K-Means Clustering, a straightforward and user-friendly algorithm, is highly effective in distinguishing sapodilla fruit maturity levels.

Scanning Micro X-ray Fluorescence and Multispectral Imaging Fusion: A Case Study on Postage Stamps.

Scanning micrο X-ray fluorescence (μ-XRF) and multispectral imaging (MSI) were applied to study philately stamps, selected for their small size and intricate structures. The μ-XRF measurements were accomplished using the M6 Jetstream Bruker scanner under optimized conditions for spatial resolution, while the MSI measurements were performed employing the XpeCAM-X02 camera. The datasets were acquired asynchronously. Elemental distribution maps can be extracted from the μ-XRF dataset, while chemical distribution maps can be obtained from the analysis of the multispectral dataset. The objective of the present work is the fusion of the datasets from the two spectral imaging modalities. An algorithmic co-registration of the two datasets is applied as a first step, aiming to align the multispectral and μ-XRF images and to adapt to the pixel sizes, as small as a few tens of micrometers. The dataset fusion is accomplished by applying k-means clustering of the multispectral dataset, attributing a representative spectrum to each pixel, and defining the multispectral clusters. Subsequently, the μ-XRF dataset within a specific multispectral cluster is analyzed by evaluating the mean XRF spectrum and performing k-means sub-clustering of the μ-XRF dataset, allowing the differentiation of areas with variable elemental composition within the multispectral cluster. The data fusion approach proves its validity and strength in the context of philately stamps. We demonstrate that the fusion of two spectral imaging modalities enhances their analytical capabilities significantly. The spectral analysis of pixels within clusters can provide more information than analyzing the same pixels as part of the entire dataset.

Exploiting Frequency Characteristics for Boosting the Invisibility of Adversarial Attacks

Mainstream transferable adversarial attacks tend to introduce noticeable artifacts into the generated adversarial examples, which will impair the invisibility of adversarial perturbation and make these attacks less practical in real-world scenarios. To deal with this problem, in this paper, we propose a novel black-box adversarial attack method that can significantly improve the invisibility of adversarial examples. We analyze the sensitivity of a deep neural network in the frequency domain and take into account the characteristics of the human visual system in order to quantify the contribution of each frequency component in adversarial perturbation. Then, we collect a set of candidate frequency components that are insensitive to the human visual system by applying K-means clustering and we propose a joint loss function during the generation of adversarial examples, limiting the frequency distribution of perturbations during attacks. The experimental results show that the proposed method significantly outperforms existing transferable black-box adversarial attack methods in terms of invisibility, which verifies the superiority, applicability and potential of this work.

K-Means Clustering Segmentation Model Using SPSS: A Case Study at Komuka Cafe Customers

Post pandemic has led to a significant and rapid increase in the growth of the cafe business. Every year, many new cafes emerge with different and unique concepts, intensifying competition in the cafe industry and necessitating continuous innovation for cafe entrepreneurs. Previously, Komuka has always highly recommended Google Reviews to prospective customers. This has caused Komuka's market share to stagnate and not experience meaningful growth. This research aims to enhance the marketing strategy of Komuka Cafe by applying K-Means cluster analysis for customer segmentation. Through data collection from Komuka's Cafe customers and klaster analysis using SPSS, the study identifies three main customer klasters: Single Employee, Married Employee, and Single Student. The detailed marketing strategy focuses on the Single Employee group because this cluster has the lowest assessment of Komuka Cafe. In this research, the author uses the Marketing Mix model 6Ps as a framework analysis to make strategy suggestions that need to be maintained and developed including Product, Place, Promotion, People, Process, and Physical Evidence. This research provides the preferences and behavior of Single Employee groups, Komuka Cafe can improve its 6Ps marketing strategy to increase competitive advantage, customer satisfaction and loyalty, as well as increase the revenue.

A Late Meal Timing Pattern Is Associated with Insulin Resistance in European Children and Adolescents

Meal timing has been associated with metabolic markers in adults, but not in children or adolescents. The aim of this study was to investigate associations of meal timing patterns (MTPs) with insulin resistance (IR) and triglyceride levels in children and adolescents. In this cross-sectional study, we included 2,195 participants aged 8–15 years from the European I.Family study (2013/14). Habitual diet exposures were derived using 24-hr dietary recalls and HOMA-IR, HbA1c, and triglycerides were used as metabolic outcome variables. We applied k-means cluster analysis on five dietary exposures (energy proportion in the morning and evening, eating window, pre-sleep fasting and eating frequency), which revealed the following three MTPs: “early-often”, “late-long” and “late-infrequent-short”. We used linear mixed models to estimate the associations between MTPs and the z-scores of the metabolic outcome variables. The association analysis revealed differences between MTPs in HOMA-IR but not in HbA1c or triglyceride z-scores. The “late-infrequent-short” pattern was associated with a 0.19 (95%-CI: (0.01, 0.36)) higher HOMA-IR z-score compared to the “early-often” pattern in the model adjusted for age, BMI z-score, education, sex, country, and family membership. These findings suggest that the timing of meals may influence IR already in childhood and adolescence. Therefore, the time of meals should be considered in future nutrition research and dietary advice for children and adolescents.

Urbanization effects on the food-water-energy nexus within ecosystem services: A case study of the Beijing-Tianjin-Hebei urban agglomeration in China

This research centers on the necessity for synchronized management of natural resources in urban agglomerations. This study utilizes the ecosystem services theory to analyze the interplay between land use and the Food-Water-Energy (F-W-E) nexus in the Beijing-Tianjin-Hebei (B-T-H) region from 2000 to 2030. Assessment of ecosystem services is conducted using InVEST models, which include Habitat Quality (HQ), Water Yield (WY), Carbon Sequestration (CS), Soil Retention (SDR), and Food Production (FP). The findings indicate an annual increase in construction land with a concurrent notable decrease in cultivated land. Furthermore, HQ, CS, and per capita FP show an annual decline until 2020, which is expected to continue until 2030. Conversely, WY and SDR have been growing annually, albeit projected to decline by 2030. Spearman coefficient analysis uncovers synergies between HQ and CS, SDR and CS, and SDR and HQ, alongside trade-offs between CS and WY and HQ and WY. Trade-offs are also observed between FP and SDR, CS, and HQ. Applying K-means clustering analysis facilitates county-based spatial planning for the F-W-E system, providing crucial insights and suggestions for sustainable resource management.

Identifying adolescents at risk for suboptimal adherence to tuberculosis treatment: A prospective cohort study.

Adolescents account for an estimated 800,000 incident tuberculosis (TB) cases annually and are at risk for suboptimal adherence to TB treatment. Most studies of adolescent TB treatment adherence have used surveillance data with limited psychosocial information. This prospective cohort study aimed to identify risk factors for suboptimal adherence to rifampicin-susceptible TB treatment among adolescents (10-19 years old) in Lima, Peru. We collected psychosocial data using self-administered surveys and clinical data via medical record abstraction. Applying k-means cluster analysis, we grouped participants by psychosocial characteristics hypothesized to impact adherence. Then, we conducted mixed effects regression to compare suboptimal adherence-defined as <90% (missing >10% of doses)-between clusters. Treatment setting (facility vs. home) and drug formulation (single drug vs. fixed dose combination) were interaction terms. Of 249 participants, 90 (36.1%) were female. Median age was 17 (IQR: 15, 16.6) years. We identified three clusters-A, B, and C-of participants based on psychosocial characteristics. Cluster C had the lowest support from caregivers, other family members, and friends; had the weakest motivation to complete TB treatment; were least likely to live with their mothers; and had experienced the most childhood adversity. Among the 118 (47.4%) participants who received facility-based treatment with single drug formulations, adherence did not differ between Clusters A and B, but Cluster C had six-fold odds of suboptimal adherence compared to Cluster A. In Clusters B and C, adherence worsened over time, but only in Cluster C did mean adherence fall below 90% within six months. Our findings have implications for the care of adolescents with TB. When caring for adolescents with low social support and other risk factors, clinicians should take extra measures to reinforce adherence, such as identifying a community health worker or peer to provide treatment support. Implementing newly recommended shorter regimens also may facilitate adherence.

RNA-seq analysis reveals modulation of inflammatory pathways by an enriched-triterpene natural extract in mouse and human macrophage cell lines

Chronic inflammation is crucial in developing insulin resistance and type 2 diabetes. Previous studies have shown that a leaf extract of Eucalyptus tereticornis, with ursolic acid (UA), oleanolic acid (OA), and ursolic acid lactone (UAL) as the main molecules (78 %) mixed with unknown minor metabolites (22 %), provided superior anti-inflammatory, hypoglycemic, and hypolipidemic effects than reconstituted triterpenoid mixtures in macrophage cell lines and a pre-diabetic mouse model. Further identification of the molecular mechanisms of action of this mixture of triterpenes is required. This study aims to analyse the RNA expression profiles of mouse and human macrophage cell lines treated with the natural extract and its components. Activated macrophage cell lines were treated with the natural extract, UA, OA, UAL or a triterpene mixture (M1). RNA was extracted and sequenced using the DNBseq platform and the EnrichR software to perform gene enrichment analysis using the Gene Ontology database, Kyoto Encyclopedia of Genes and Genomes, and Reactome. To conduct clustering analysis, we standardised the normalised counts of each gene and applied k-means clustering. The combination of molecules in the natural extract has an additive or synergic effect that affects the expression of up-regulated genes by macrophage activation. Triterpenes (M1) regulated 76 % of human and 68 % of mouse genes, while uncharacterised minority molecules could regulate 24 % of human and 32 % of mouse genes. The extract inhibited the expression of many cytokines (IL6, IL1, OSM), chemokines (CXCL3), inflammatory mediators (MMP8 and MMP13) and the JAK-STAT signalling pathway in both models. The natural extract has a more powerful immunomodulatory effect than the triterpene mixture, increasing the number of genes regulated in mouse and human models. Our study shows that Eucalyptus tereticornis extract is a promising option for breaking the link between inflammation and insulin resistance.

VHDLISTE: An Instructional Strategy to Cover the Additional Content beyond the Curriculum

Abstract: In current scenario of engineering education, Students must have more practical knowledge and skills required to work in the technologies that are currently used in industries though they are not the part of the curriculum. The instruction based videos help students to develop their hands on the new skills that are essential for their adaptability to new technologies. Handout is the guided document which contains the instructions to follow during the laboratory session to complete the problem statement given in the assignment. Assessment of their knowledge acquirement and skills is done through grading of assignments given to them. So in this article, VHDLISTE (Instructional Videos followed by Handout Learning – Implement, Submit, Test, and Evaluate) instructional strategy is designed and implemented for Tableau: A Visual Data Analytics Tool which is the part of the course Business Intelligence and Big Data Analytics of Final Year Computer Science and Engineering but is not covered in the syllabus. This VHDLISTE instructional strategy consists of six phases V- Instructional Videos, HDL- Handout Learning, I – Implement, S – Submit, T-Test, and E – Evaluate. In first phase V - Instructional Videos phase, students watch the instructional videos prepared by the instructor to get acquainted with the Tableau tool. In second phase HDL, students go through the lab handouts to read the instructions given for completing the lab assignment. In I – Implement phase, students implement the problem statement related to Tableau tool while test on covered topics is conducted in T-Test phase. In S – Submit, students submit the problem statement on institutional MOODLE. Instructor conducts the test on covered topic in T-Test phase. Instructor evaluates the problem statement as well as test submitted by students on institutional MOODLE in E – Evaluate phase.Learning objective (LOs) of this study is to teach the problem analysing and solving ability. These LOs are 1) Apply Filters on Dimensions and Measures for given dataset using Tableau (LO1) and 2) Apply K-means Clustering on given dataset in Tableau (LO2). The research question is - Whether the use of instructional strategy VHDLISTE is useful for learning the content beyond the curriculum? This activity is conducted for two Academic Years 2020-21, 2021-22 and 2022-23. The result of applying this strategy and perception of students are also explained. The result showed that this strategy improves the problem analysis as well as problem solving ability of students. Using t-test statistical analysis, it is found that there is significant difference in mean score between sample and overall population means for LO1 as well as LO2 for all these three years. Keywords: Learning Objectives, Handouts, Instruction based Videos, t-Test

Specificity of carbon nanotube accumulation and distribution in cancer cells revealed by K-means clustering and principal component analysis of Raman spectra.

Single-walled carbon nanotubes (SWCNTs) show great potential for their application as cancer therapeutic nanodrugs, but the efficiency and mechanism of their accumulation in the cell, the modulation of cell activity, and the strong dependence of the results on the type of capping molecule still hinder the transfer of SWCNTs to the clinic. In the present study, we determined the mechanism and sequence of accumulation, distribution and type discrimination of SWCNTs in glioma cells by applying K-means clustering and principal component analysis (PCA) of Raman spectra of cells exposed to SWCNTs capped with either DNA or oligonucleotides (ON). Based on the specific biochemical information uncovered by PCA and further applied to K-means, we show that the accumulation of SWCNT-DNA occurs in two phases. The first phase involves the transport of SWCNT-DNA through vesicles and its redistribution in the cytoplasm, which is reflected in two SWCNT-related clusters. The second phase begins after 18 hours of interaction between cells and SWCNT-DNA. PCA shows the appearance of two SWCNT-associated PC loadings, reflected by the addition of a new cluster of SWCNTs with a narrowed and shifted G-peak in the spectra. It is caused by the loss of DNA capping and clumping of SWCNTs and triggered by the acidic conditions in autolysosomes resulting from the fusion of transport vesicles with lysosomes. SWCNTs penetrate all cellular compartments after 42-66 hours and lead to cell death. The clumped SWCNTs are released to the outside. In contrast, SWCNT-ON is hardly accumulated in glioma cells and after 72 hours of exposure to SWCNT-ON, the accumulation of SWCNTs corresponds to the first stage without reaching the second. PCA made it possible to separate the characteristics of cellular components against the high-intensity Raman signal from nanotubes and, thus, to propose the mechanism of accumulation and metabolism of nanomaterials in living cells without the use of additional research approaches. Our results elucidate the time dependence of the accumulation of SWCNTs on the capping molecule. We expect that our results can make an important contribution to the use of these nanomaterials in the clinic.

Application of PCA and K-Means Clustering Methods to Identify Diabetes Mellitus Patient Groups Based on Risk Factors

Diabetes mellitus is a chronic disease characterized by high levels of glucose (sugar) in the blood that is high for a long period of time. Identification is the process of recognizing and determining the characteristics of a particular object or entity. hypertension (high blood pressure), smoking and lack of physical activity can affect the condition of diabetes mellitus patients. Therefore, an approach is needed that can identify groups of diabetic patients based on their risk factors, so that appropriate management and treatment can be carried out. The purpose of this study is to apply PCA method by reducing data dimension to identify the linear combination of the most contributing risk factors in diabetes mellitus patient data and apply K-Means Clustering to cluster into groups based on similar risk factors. The methods to be used are Principal Component Analysis (PCA) and K-Means Clustering. type of quantitative research, this research can be categorized as analytic research, variables are risk factors for diabetes mellitus disease. The results of research using the PCA (principal component analysis) method obtained 9 main components (PC) 86.9275%. correlation between attributes and principal components, then a matrix component is formed with a loading value that the greater the value, the stronger the correlation with the principal component formed with a cut off point of loading value&gt; 0.4 regardless of positive and negative. By using the K-Means Clustering method, The clustering results obtained are divided into 3 groups of diabetes patients based on existing risk factors. Centroid C1 represents a group of diabetes mellitus patients whose condition is at a mild level, while Centroid C2 represents a group of diabetes mellitus patients who are at a moderate level, and Centroid C3 represents a group of patients with severe or dangerous diabetes mellitus.

Neuroaffective reactivity profiles are associated with vulnerability to e-cigarette use

BackgroundWe tested whether neuroaffective responses to motivationally salient stimuli are associated with vulnerability to cue-induced e-cigarette use in e-cigarette naïve adults who smoke daily. We hypothesized that individuals with stronger neuroaffective responses to nicotine-related cues than to pleasant stimuli (the C>P reactivity profile) would be more vulnerable to cue-induced nicotine self-administration than individuals with stronger neuroaffective responses to pleasant stimuli than to nicotine-related cues (the P>C reactivity profile). MethodsWe used event-related potentials (ERPs, a direct measure of cortical activity) to measure neuroaffective reactivity to pleasant, unpleasant, neutral, and nicotine-related cues indicating the opportunity to use an e-cigarette in 36 participants. For each picture category, we computed the amplitude of the late positive potential (LPP), a robust index of motivational salience. To identify each individual’s neuroaffective reactivity profile we applied k-means cluster analysis on the LPP responses. We compared the e-cigarette use frequency across profiles using quantile regression for counts. ResultsK-means cluster analysis assigned 18 participants to the C>P profile and 18 participants to the P>C profile. Individuals with the C>P neuroaffective profile used the e-cigarette significantly more often than those with the P>C profile. Significant differences in the number of puffs persisted across different quantiles. ConclusionsThese results support the hypothesis that individual differences in the tendency to attribute motivational salience to drug-related cues underlie vulnerability to cue-induced drug self-administration. Targeting the neuroaffective profiles that we identified with tailored treatments could improve clinical outcomes.

Analisis Karakteristik Keberadaan Perbankan di Nusa Tenggara Barat Terhadap Kondisi Perekonomian Daerah Menggunakan K-Means Clustering

In certain areas, there are still many people who have to travel long distances to access some banks. Difficult mobility is considered to hinder business activities. The West Nusa Tenggara (NTB) Province is one of the favorite travel destinations for some foreigner tourists as well as domestic tourists because of its natural beauty and cultural diversity. the existence of some banks in the NTB Province , is important to facilitate the circulation of money. For this reason, this study aims to analyze the existence of some banks in the NTB Province and the condition of mobility in accessing themto regional economic conditions by applying K-Means clustering. Our results show that there are two clusters, , where the cluster 2 is an urban area and a tourist area. It has charactersitic has a GDP greater than cluster 1.

Meteorological and chemical controls on surface ozone diurnal variability in Beijing: A clustering-based perspective

This study aimed to reveal the dominant, recurring patterns of surface ozone diurnal cycles in Beijing, their quantitative characteristics, namely, occurrence frequency and transition probability, and their potential linkages to meteorological and chemical processes based on multi-source observations (surface ozone, satellite-based ozone precursors, ground-based and sonde-based meteorological measurements) and reanalyses (ERA5 meteorological and TCR2 chemical reanalyses). By applying K-means clustering, four distinct ozone diurnal patterns (Cluster 1: all-day low anomaly; Cluster 2: nighttime high-daytime low anomaly; Cluster 3: nighttime low-daytime high anomaly; Cluster 4: extreme pollution pattern with all-day high anomaly) are objectively identified from 2013 to 2020 warm-season (May–September) surface ozone measurements. Despite a relatively low total occurrence, the extreme pollution pattern tended to persist over an extended period (average persistence of 2.3 days). Composite analyses of multi-source meteorology indicate that the elevated nocturnal ozone in Cluster 2 is associated with the compensating downdraft flow of nocturnal convective storms. By contrast, elevated nocturnal ozone in Cluster 4 was associated with the prevailing low-level jet (LLJ) over the whole North China Plain (NCP) region. The LLJ-related dynamical processes trigger regional transport in the residual layer and downward mixing from the residual layer, elevating the background ozone level in downstream Beijing. Moreover, Cluster 4 is under the control of 500-hPa anomalous high pressure, which leads to daytime hot-dry weather favoring the occurrence of extreme afternoon pollution via photochemical ozone production. From Cluster 1 to Cluster 4, satellite-based formaldehyde (CH2O) increased by 71% accompanied by less change in nitrogen dioxide (NO2), causing a shift in ozone production sensitivity. Because extreme pollution pattern typically occurs in the transition regime, synergetic emission control of volatile organic compounds (VOCs) and nitrogen oxides (NOx) would be effective in alleviating severe ozone pollution in Beijing.

A person-centered perspective on the combined DSM-5 AMPD/ICD-11 personality model: Utility, relationship with the categorical personality disorder model, and capacity to differentiate between levels of identity functioning.

Both the ICD-11 classification of Personality Disorders and the DSM-5 Alternative Model for Personality Disorders (DSM-5 AMPD) conceptualize personality pathology in a dimensional way, but differ in the way they carve up their respective pathological personality domains. Recently, a combination of ICD-11 and DSM-5 AMPD descriptive pathological personality traits, the Modified Personality Inventory for DSM-5-Brief Form Plus (PID5BF + M), was developed. We investigated the utility of the additional ANANKASTIA domain (not represented in the DSM-5 AMPD) as well as of the additional PSYCHOTICISM domain (not represented in the ICD-11 model) in the identification of meaningful pathological personality domain clusters based on the PID5BF + M. Next to the classical 2- and 3-cluster solutions, we examined whether the presence of the additional ANANKASTIA domain would also gave rise to a meaningful 4-cluster solution. We then validated these clusters by investigating differences between them in mean DSM-5 Section II cluster A, B, and C personality disorder scores. Finally, we investigated whether cluster membership was able to differentiate between levels of identity functioning, a key feature of personality disorder severity in both the ICD-11 model and the DSM-5 AMPD. We used a Flemish community sample of 242 participants, and applied k-means cluster analyses in a two-step manner on PID5BF + M domains to investigate 2-, 3-, and 4-cluster solutions. We used MANOVAs to examine differences between clusters in PID5BF + M domains, DSM-IV/DSM-5 Section II Assessment of Personality disorders (ADP-IV) cluster A, B, and C scores, and Self-Concept and Identity Measure (SCIM) scores. Cluster analyses on PID5BF + M pathological personality domains (1) revealed meaningful 2-, 3-, and 4-cluster solutions, with the 4-cluster solution explaining the most variance in the clustering variables, (2) allowed to identify a classical Overcontrolled cluster which DSM-5 AMPD PID-5 does not, and (3) demonstrated the utility of representing ANANKASTIA and DISINHIBITON as separate pathological personality domains. PID5BF + M clusters (5) were informative of DSM-5 Section II cluster A, B, and C personality disorder scores and (6) showed different levels of clinical-developmental Identity functioning. Current results demonstrate the utility of a combined ICD-11/DSM-5 AMPD view from a person-centered perspective.

Deep convolutional correlation iterative particle filter for visual tracking

This work proposes a novel framework for visual tracking based on the integration of an iterative particle filter, a deep convolutional neural network, and a correlation filter. The iterative particle filter enables the particles to correct themselves and converge to the correct target position. We employ a novel strategy to assess the likelihood of the particles after the iterations by applying K-means clustering. Our approach ensures a consistent support for the posterior distribution. Thus, we do not need to perform resampling at every video frame, improving the utilization of prior distribution information. Experimental results on three different benchmark datasets show that our tracker performs favorably against state-of-the-art methods.