Probabilistic Clustering Research Articles

A Novel Bayesian probabilistic distance clustering algorithm

A Novel Bayesian probabilistic distance clustering algorithm

Structured multi-view k-means clustering

K-means is a very efficient clustering method and many multi-view k-means clustering methods have been proposed for multi-view clustering during the past decade. However, since k-means have trouble uncovering clusters of varying sizes and densities, these methods suffer from the same performance issues as k-means. Improving the clustering performance of multi-view k-means has become a challenging problem. In this paper, we propose a new multi-view k-means clustering method that is able to uncover clusters in arbitrary sizes and densities. The new method simultaneously performs three tasks, i.e., sparse connection probability matrices learning, prototypes aligning, and cluster structure learning. We evaluate the proposed new method by 5 benchmark datasets and compare it with 11 multi-view clustering methods. The experimental results on both synthetic and real-world experiments show the superiority of our proposed method.

Transitive reasoning: A high-performance computing model for significant pattern discovery in cognitive IoT sensor network

Current research on the Internet of Things (IoT) has given rise to a new field of study called cognitive IoT (CIoT), which aims to incorporate cognition into the designs of IoT systems. Consequently, the CIoT inherits specific attributes and challenges from IoT. The CIoT applications generate vast, diverse, constantly changing, and time-dependent data due to the billions of devices involved. The efficient operation of these CIoT systems requires the extraction of valuable insights from vast data sources in a computationally efficient manner. Therefore, this study proposes transitive reasoning to glean significant concepts and patterns from a 21.25-year environmental dataset. To reduce the effects of missing entries, the proposed methodology includes a grouping of data using probabilistic clustering and applying total variance regularization in the alternate direction method of multipliers (ADMM) to regularize the sensory data. As a result, noisy entries will be less conspicuous. Afterward, it calculates the transitional plausibility value for each cluster using the transited value and then turns it into binary data to create concept lattices. In addition, each concept that is formed is assigned a weight, and the concept with the largest transitive strength value is chosen, followed by calculating the mean value. Therefore, this pattern is seen as significant. Experimental results on 21.25-year environmental data show an accuracy of over 99.5%, outperforming competing methods, as shown by cross-validation using multiple metrics.

Augmented ELBO regularization for enhanced clustering in variational autoencoders

With significant advances in deep neural networks, various new algorithms have emerged that effectively model latent structures within data, surpassing traditional clustering methods. Each data point is expected to belong to a single cluster in a typical clustering algorithm. However, clustering based on variational autoencoders (VAEs) represents the expectation of the overall clusters, denoted as c=1,…,K in the KL divergence term. Consequently, the latent embedding z can be learned to exist across multiple clusters with relatively balanced probabilities, rather than being strongly associated with a specific cluster. This study introduces an additional regularizer to encourage the latent embedding z to have a strong affiliation with specific clusters. We introduce optimization methods to maximize the ELBO that includes the newly added regularization term and explore methods to eliminate computationally challenging terms. The positive impact of this regularization on clustering accuracy was verified by examining the variance of the final cluster probabilities. Furthermore, an enhancement in the clustering performance was observed when regularization was introduced.

Remote Sensing Image Segmentation Based on Probabilistic Fuzzy Local Information Clustering

Abstract. In this paper, a remote sensing image segmentation based on probabilistic fuzzy local information clustering algorithm is proposed. First, assuming that the spectral measure within the same ground object follows the same probability distribution. The dissimilarity between pixel and object is modeled by the negative logarithm of the Gaussian probability density function. It can improve the noise sensitive problem caused by the Euclidean distance which can only describe the data with isotropic distribution. Then, in order to consider the effect of local spatial constraint, on the one hand, the probability measure is used to modify the local fuzzy factor to establish the dissimilarity measure with spatial constraints. On the other hand, the hidden Markov random field is used to model the prior probability model of pixel membership. Next, the entropy regularization term of the objective function is built by combining the Kullback-Leibler(KL) maximum entropy model to further improve the robustness and noise resistance. The qualitative and quantitative analysis of simulated image and different types of real remote sensing images show that the proposed algorithm can effectively overcome the above problems and further improve the accuracy of image segmentation to over 95%.

Prevalence and correlates of mental disorders among women: results from the Saudi National Mental Health Survey

BackgroundMental disorders account for a significant proportion of the world’s disease burden and are more significant among females than males. However, most global mental health research is sex neutral, including in the Kingdom of Saudi Arabia. This study, therefore, estimated the prevalence of mental disorders and investigated the sociodemographic correlates, sex disadvantage factors, and treatment-seeking in Saudi women concerning lifetime and 12-month mental disorders.MethodThe Saudi National Mental Health Survey is a stratified multistage clustered area probability design. Lifetime and 12-month mental disorders were assessed through the Composite International Diagnostic Interview (CIDI 3.0). The correlates considered for this study included age-at-interview, education, marital status, employment status, socioeconomic status (SES), any chronic condition and household characteristics (region, urbanicity, and income), as well as domestic violence, age at marriage and in a polygamous marriage. Data was analysed using PROC SURVEYFREQ procedure as well as logistic regression in SAS 9.2.ResultsOverall, 24.7% and 35.9% of Saudi women experienced at least one of the disorders in the prior 12 months and at least once in their lifetime, respectively. Anxiety disorders were the most frequently reported 12-month and lifetime disorders, followed by mood disorders. The region, urbanicity, chronic conditions, employment status, as well as certain sex disadvantage factors were significantly associated with both 12-month and lifetime disorders. Most women did not seek treatment for 12-month mental disorders (86.2%) and lifetime disorders (73.8%).ConclusionOur study confirms that mental health issues, particularly anxiety and mood disorders, are highly prevalent among Saudi women, influenced by chronic conditions and sex-related factors like domestic violence and polygamy. Future research should focus on improving mental health literacy, using rigorous study designs to explore female-specific variables, and investigating genetic and environmental factors.

Photometric and Kinematic Studies of Open Clusters Ruprecht 1 and Ruprecht 171

ABSTRACTThis study outlines a detailed investigation using CCD UBV and Gaia DR3 data sets of the two open clusters Ruprecht 1 (Rup‐1) and Ruprecht 171 (Rup‐171). Fundamental astrophysical parameters such as color excesses, photometric metallicities, ages, and isochrone distances were based on UBV‐data analyses, whereas membership probability calculations, structural and astrophysical parameters, as well as the kinematic analyses were based on Gaia DR3‐data. We identified 74 and 596 stars as the most probable cluster members with membership probabilities over 50% for Rup‐1 and Rup‐171, respectively. The color excesses were obtained as and mag for Rup‐1 and Rup‐171, respectively. Photometric metallicity analyses were performed by considering F‐G type main‐sequence member stars and found to be and dex for Rup‐1 and Rup‐171, respectively. Ages and distances were based on both UBV and Gaia‐data analyses; according to isochrone‐fitting these values were estimated to be Myr, pc for Rup‐1 and Myr, pc for Rup‐171. The present‐day mass function slope of Rup‐1 was estimated as and Rup‐171 as . Galactic orbit integration analyses showed that both of the clusters might be formed outside the solar circle.

Unresolved binary systems with white dwarfs in open star clusters

We invesigate unresolved binary systems with components of main sequence star (MS) and white dwarf (WD) in nine open clusters. These systems are located below and to the left of the main sequence at the colour-magnitude diagram. We compare the number of cluster stars, which have likely evolved into white dwarfs, with the number of candidates for unresolved binary systems with WD. The number of probable cluster members, lying below the main sequence, is generally less than the expected number of white dwarfs. The observations in the ultraviolet could detect WDs and unresolved binary WD+MS systems more confidently than the observations in the visible range.

Unveiling Subarcsecond Multiplicity in the Pleiades with Gaia Multicolor Photometry

The list of 409 probable cluster members down to G = 15mag (m ≳ 0.5M ⊙) is compiled for the two degree radius of the Pleiades, based on astrometric data from Gaia DR3 and the PPMXL catalog, along with several radial velocity surveys, including APOGEE and LAMOST. This approach allows for the inclusion of binary stars with unreliable Gaia solutions, thereby eliminating associated bias. Thus, the often-neglected 14 sources with Gaia two-parameter solutions are included. The subsequent analysis of color–magnitude and color–color diagrams exploits artifacts in Gaia photometric data, caused by the different field sizes used to measure fluxes in the G, B p , and R p passbands, to reveal binary stars with subarcsecond angular separation. The findings are validated with prior high-resolution observations. Overall, 24 ± 3 cluster members with angular separation between 0.″1 and 1″ (13.5–135 au projected distance) and mass ratio q > 0.5 are deemed binary, indicating a binarity fraction of 6 ± 1%.

Nationally Representative Rates of Incident Prescription Opioid Use Among United States Adults and Selected Subpopulations: Longitudinal Cohort Study From the National Health Interview Survey, 2019 to 2020

Nationally representative rates of incident prescription opioid use in the United States adult population and selected subpopulations are unknown. Using the National Health Interview Survey (2019–2020) longitudinal cohort, a cohort with 1-year follow-up created using random cluster probability sampling of noninstitutionalized civilian U.S. adults, we estimated rates and predictors of incident opioid use. Of 21,161 baseline (2019) participants randomly chosen for follow-up, the final analytic sample included 10,415 who also participated in 2020. Exposure variables were selected per the socio-behavioral model of health care utilization: predisposing characteristics (sex, age, race, etc), enabling characteristics (socioeconomic status, insurance status), health status (pain, disability, comorbidities, etc), and health care use (office visits, emergency room visits, and hospitalizations). Among adults who did not use prescription opioids in 2019, a 1-year cumulative incidence of 4.1% (95% confidence interval [CI]: 3.5–4.6) was seen in 2020, with an incidence rate (IR) of 32.6 cases of new prescription opioid use per 1,000 person-years (PYs). Cumulative incidence, IR, and adjusted relative risk (RR) varied by participant characteristics. We observed the highest IR in those with ineffective pain treatment (81.6 cases per 1,000 PY) and those who visited the emergency room ≥3 times (93.8 cases per 1,000 PY). Participants reporting ≥4 painful conditions had an adjusted RR of 2.9 (95% CI: 2.0–4.1), while the RR for those with sleep problems was 2.3 (95% CI: 1.7–3.1). Overall, this study presents nationally representative rates of incident prescription opioid use and suggests that some participants are using prescription opioids as an early-resort analgesic contrary to best-practice guidelines. PerspectiveThis longitudinal cohort study presents nationally representative rates of incident prescription opioid use in U.S. adults and selected subpopulations. Our data suggest that some participants are using prescription opioids as a first-line or early-resort analgesic, contrary to best-practice guidelines.

THE EFFECT OF WORK DISCIPLINE AND MOTIVATION ON PERFORMANCE EMPLOYEES IN THE POLICE CIVIL SERVICE UNIT OF KARAWANG REGENCY

Purpose of the study — he research aims to determine employee performance which can be influenced simultaneously and partially by the work discipline and motivation of the Karawang Regency Satpol PP.Research method— he population in this study was 108 employees who had a tolerable error rate of 10% or an accuracy rate of 90%. Therefore, a purposive cluster probability sampling technique was used to sample 52 employees to represent the population. Descriptive and verification methods are used in this research, which means collecting data, displaying, analyzing and testing hypotheses, before making conclusions and recommendations. Descriptive statistical analysis, classical assumption tests, and multiple linear regression analysis were used with SPSS 26 hypothesis testing.Result— This research results that employee performance can be partially significantly and positively influenced by work discipline, employee performance can be partially significantly and positively influenced by work motivation, Employee performance can be affected simultaneously by discipline and work motivation.Conclusion— The work discipline and motivation of employees in the Karawang Regency Satpol PP have a significant and positive effect on employee performance, both partially and simultaneously.

Generalized load modeling approach considering multiple distributed generation integration

To solve the problem of node characteristic uncertainty caused by the connection of multiple distributed generations to the original load node, a new method for modeling the steady-state characteristics of generalized load nodes based on scene clustering was proposed. Scenario clustering takes into account three factors: wind farm output fluctuation, photovoltaic output fluctuation, and demand-side power load fluctuation, and extracts typical scenarios of generalized load data. According to the uncertainty change of node characteristics in different scenarios, the clustering data is segmented and refined, and its probability distribution is counted. Clustering information, probability information, and maximum value are used as three types of features of generalized load and are custom encoded as 12-dimensional feature vectors as neural network inputs. To improve the identification performance of generalized load features, the back propagation (BP) dichotomy neural network is used to build a unified model of node features. The simulation results show that the proposed method can not only model accurately but also introduce probability information through statistical data samples, which can provide an auxiliary reference for the prediction and regulation of generalized loads.

Detrital muscovite 40Ar/39Ar geochronology and provenance of Cenozoic deposits in the Qaidam basin, northern Tibetan Plateau and comparison with detrital zircon and apatite records

The single-grain detrital multi-mineral dating techniques are powerful tools for tracing sediment provenance. However, natural biases in sediment source-to-sink systems may lead to heterogeneous sink signals based on detrital single-grain ages. Differing provenance interpretations may arise from detrital multi-mineral geochronologic data because of the potentially diverse origins and transport behaviors of the analyzed detrital minerals. To test this hypothesis, we present new single-grain detrital muscovite 40Ar/39Ar dating results from the Cenozoic Qaidam basin and compile published detrital muscovite 40Ar/39Ar, detrital zircon U-Pb, and detrital apatite fission track age data to interpret muscovite provenance and to explore the potential inconsistencies among multi-dating results and related controlling mechanisms. Our results indicate that detrital muscovite grains from most Cenozoic deposits have a similar, dominated 300–400 Ma age range, with subordinate 200–300 Ma and 400–500 Ma. The new and compiled 40Ar/39Ar age data demonstrate that detrital muscovite records from the Cenozoic Qaidam basin are stratigraphically variable. Samples from middle fine-grained deposits-dominated strata, representing the paleo-megalake period, exhibit a relatively wider age range compared to those from lower and upper coarse-grained deposits-dominated strata, indicating a probable control of sedimentary environments. Comparisons of detrital muscovite, zircon, and apatite records indicate that the detrital zircon age spectra have the oldest ages, the widest age ranges, and the largest number of age groups. Detrital zircon and apatite age spectra exhibit higher spatial variations than detrital muscovite records, likely due to the distinct source-to-sink behaviors of these minerals. The magmatic, metamorphic, and uplift history of source terranes, along with system closure temperatures, originally determine detrital mineral age records, i.e., high-temperature systems yield older ages, and vice versa. Parent-rock types, mineral fertility, and weathering resistance control the numbers and probabilities of the detrital age clusters. Differences in mineral shape, density, and transport and depositional processes (i.e., different transport loads of light, platy minerals vs. heavy, granular minerals) may cause variations in age spectra. These differences in detrital mineral behaviors warrant more attention in multi-mineral dating-based sediment provenance studies. Additionally, 40Ar/39Ar age data compilation based on global Cenozoic sedimentary basins and modern rivers indicates that detrital muscovite records have potentials to determine tectonic settings, but climate-induced exhumation signals need to be distinguished before interpretation.

PPFCM-SMOTE: a novel balancing system for anomaly detection in IoT edge using probabilistic possibilistic fuzzy clustering and SMOTE

PPFCM-SMOTE: a novel balancing system for anomaly detection in IoT edge using probabilistic possibilistic fuzzy clustering and SMOTE

Cluster-based Graph Collaborative Filtering

Graph Convolution Networks (GCNs) have significantly succeeded in learning user and item representations for recommendation systems. The core of their efficacy is the ability to explicitly exploit the collaborative signals from both the first- and high-order neighboring nodes. However, most existing GCN-based methods overlook the multiple interests of users while performing high-order graph convolution. Thus, the noisy information from unreliable neighbor nodes ( \(e.g.\) , users with dissimilar interests) negatively impacts the representation learning of the target node. Additionally, conducting graph convolution operations without differentiating high-order neighbors suffers the over-smoothing issue when stacking more layers, resulting in performance degradation. In this paper, we aim to capture more valuable information from high-order neighboring nodes while avoiding noise for better representation learning of the target node. To achieve this goal, we propose a novel GCN-based recommendation model, termed Cluster-based Graph Collaborative Filtering (ClusterGCF). This model performs high-order graph convolution on cluster-specific graphs, which are constructed by capturing the multiple interests of users and identifying the common interests among them. Specifically, we design an unsupervised and optimizable soft node clustering approach to classify user and item nodes into multiple clusters. Based on the soft node clustering results and the topology of the user-item interaction graph, we assign the nodes with probabilities for different clusters to construct the cluster-specific graphs. To evaluate the effectiveness of ClusterGCF, we conducted extensive experiments on four publicly available datasets. Experimental results demonstrate that our model can significantly improve recommendation performance.

The role of heterogeneous stress in earthquake cycle models of the Hikurangi-Kermadec subduction zone

Summary Seismic and tsunami hazard modelling and preparedness are challenged by uncertainties in the earthquake source process. Important parameters such as the recurrence interval of earthquakes of a given magnitude at a particular location, the probability of multi-fault rupture, earthquake clustering, rupture directivity and slip distribution are often poorly constrained. Physics-based earthquake simulators, such as RSQSim, offer a means of probing uncertainties in these parameters by generating long-term catalogues of earthquake ruptures on a system of known faults. The fault initial stress state in these simulations is typically prescribed as a single uniform value, which can promote characteristic earthquake behaviours and reduce variability in modelled events. Here, we test the role of spatial heterogeneity in the distribution of the initial stresses and frictional properties on earthquake cycle simulations. We focus on the Hikurangi-Kermadec subduction zone, which may produce Mw > 9.0 earthquakes and likely poses a major hazard and risk to Aotearoa New Zealand. We explore RSQSim simulations of Hikurangi-Kermadec subduction earthquake cycles in which we vary the rate and state coefficients (a and b). The results are compared with the magnitude-frequency distribution (MFD) of the instrumental earthquake catalogue and with empirical slip scaling laws from global earthquakes. Our results suggest stress heterogeneity produces more realistic and less characteristic synthetic catalogues, making them particularly well suited for hazard and risk assessment. We further find that the initial stress effects are dominated by the initial effective normal stresses, since the normal stresses evolve more slowly than the shear stresses. A heterogeneous stress model with a constant pore-fluid pressure ratio and a constant state coefficient (b) of 0.003 produces the best fit to MFDs and empirical scaling laws, while the model with variable frictional properties produces the best fit to earthquake depth distribution and empirical scaling laws. This model is our preferred initial stress state and frictional property settings for earthquake modelling of the Hikurangi-Kermadec subduction interface. Introducing heterogeneity of other parameters within RSQSim (e.g. friction coefficient, reference slip rate, characteristic distance, initial state variable, etc) could further improve the applicability of the synthetic earthquake catalogues to seismic hazard problems and form the focus of future research.

Correspondence of Consonant Clustering with Particular Vowels in German Dialects

Recent work found a correspondence between consonant clustering probability in monosyllabic lexemes and the three vowel types, short and long monophthong and diphthong, in German dialects. Furthermore, that correspondence was found to be bound to a North–South divide. This paper explores the preferences in consonant clustering of particular vowels by analyzing the PhonD2-Corpus, a large database of phonotactic and morphological information. The clustering probability of the diphthongs is positively correlated with frequency while the other vowels showed particular preferences that are not positively correlated with frequency. However, all of them are determined by a threefold pattern: short monophthongs prefer coda clusters, diphthongs onset clusters and long monophthong are balanced. Furthermore, it was found that this threefold pattern seems to have evolved from an originally twofold pattern (short monophthong prefers coda clusters and long monophthong and diphthong prefer onset clusters) in Middle High and Low German. This result is then further considered under the aspect of the compensation of the syllable weight and moraicity. Furthermore, some interesting parallels with the syllable vs. word-language typology framework are noted.

ATC-SD Net: Radiotelephone Communications Speaker Diarization Network

This study addresses the challenges that high-noise environments and complex multi-speaker scenarios present in civil aviation radio communications. A novel radiotelephone communications speaker diffraction network is developed specifically for these circumstances. To improve the precision of the speaker diarization network, three core modules are designed: voice activity detection (VAD), end-to-end speaker separation for air–ground communication (EESS), and probabilistic knowledge-based text clustering (PKTC). First, the VAD module uses attention mechanisms to separate silence from irrelevant noise, resulting in pure dialogue commands. Subsequently, the EESS module distinguishes between controllers and pilots by levying voice print differences, resulting in effective speaker segmentation. Finally, the PKTC module addresses the issue of pilot voice print ambiguity using text clustering, introducing a novel flight prior knowledge-based text-related clustering model. To achieve robust speaker diarization in multi-pilot scenarios, this model uses prior knowledge-based graph construction, radar data-based graph correction, and probabilistic optimization. This study also includes the development of the specialized ATCSPEECH dataset, which demonstrates significant performance improvements over both the AMI and ATCO2 PROJECT datasets.

Trends of diabetes in Beijing, China.

The global rise in diabetes prevalence is a pressing concern. Despite initiatives like "The Healthy Beijing Action 2020-2030" advocating for increased awareness, treatment, and control, the specific situation in Beijing remains unexplored. This study aimed to analyze the trends in diabetes prevalence, awareness, treatment, and control among Beijing adults. Through a stratified multistage probability cluster sampling method, a series of representative cross-sectional surveys were conducted in Beijing from 2005 to 2022, targeting adults aged 18-79 years. A face-to-face questionnaire, along with body measurements and laboratory tests, were administered to 111,943 participants. Data from all survey were age- and/or gender-standardized based on the 2020 Beijing census population. Annual percentage rate change (APC) or average annual percentage rate change (AAPC) was calculated to determine prevalence trends over time. Complex sampling logistic regression models were employed to explore the relationship between various characteristics and diabetes. From 2005 to 2022, the total prevalence of diabetes among Beijing adults aged 18-79 years increased from 9.6% (95% CI: 8.8-10.4%) to 13.9% (95% CI: 13.1-14.7%), with an APC/AAPC of 2.1% (95% CI: 1.1-3.2%, P<0.05). Significant increases were observed among adults aged 18-39 years and rural residents. Undiagnosed diabetes rose from 3.5% (95% CI: 3.2-4.0%) to 7.2% (95% CI: 6.6-7.9%) with an APC/AAPC of 4.1% (95% CI: 0.5-7.3%, P<0.05). However, diabetes awareness and treatment rates showed annual declines of 1.4% (95% CI: -3.0% to -0.2%, P<0.05) and 1.3% (95% CI: -2.6% to -0.2%, P<0.05), respectively. The diabetes control rate decreased from 21.5% to 19.1%, although not statistically significant (APC/AAPC=-1.5%, 95% CI: -5.6% to 1.9%). Overweight and obesity were identified as risk factors for diabetes, with ORs of 1.65 (95% CI: 1.38-1.98) and 2.48 (95% CI: 2.07-2.99), respectively. The prevalence of diabetes in Beijing has significantly increased between 2005 and 2022, particularly among young adults and rural residents. Meanwhile, there has been a concerning decrease in diabetes awareness and treatment rates, while control rates have remained stagnant. Regular blood glucose testing, especially among adults aged 18-59 years, should be warranted. Furthermore, being male, elderly, overweight, or obese was associated with higher diabetes risk, suggesting the needs for targeted management strategies.

Improving wireless sensor network lifespan with optimized clustering probabilities, improved residual energy LEACH and energy efficient LEACH for corner-positioned base stations

The goal of this paper's novel energy-conscious routing method is to optimize energy usage and extend network lifespans using a new clustering probability. Versatile arrangements and a longer network lifespan (until the last node dies) are achieved through cluster-based routing strategies. Existing algorithms, such as low energy adaptive clustering hierarchy (LEACH), residual energy LEACH (RES-EL), and distributed residual energy LEACH (DIS-RES-EL), have been compared to the newly proposed algorithms: improved residual energy LEACH (IMP-RES-EL) and energy efficient LEACH (EEL). IMP-RES-EL and EEL outperform all other stated algorithms by extending the network lifespan, enhancing stability, increasing the number of aggregated data packets transmitted from cluster heads to the base station (BS), and selecting cluster heads with energy efficiency and optimal routing within the network. The proposed approaches outperform existing algorithms, particularly when every corner-located BS is considered in the wireless sensor network (WSN). The network lifespan in rounds increased by 36 %, the number of aggregated data packets from cluster heads to the BS increased by 44 %, and the efficiency of corner-located BSs improved by 20 %. Extensive simulations on five distinct topologies were reviewed and compared to the three techniques listed above, demonstrating the superiority of the proposed algorithms.