Hierarchical Clustering Model Research Articles

Implementation of ward's agglomerative hierarchical clustering model to detect pulmonary tuberculosis endemic areas in Aceh Utara regency

This study aims to detect pulmonary Tuberculosis (TB) endemic areas in Aceh Utara Regency based on altitude, population density, and the number of TB cases. The Agglomerative Hierarchical Clustering (AHC) algorithm was used to cluster 27 subdistricts, each measured by these three factors. The clustering results divided the subdistricts into three main clusters with distinct characteristics. Cluster 1 consists of subdistricts with low altitude, high population density, and relatively high numbers of TB cases, identifying this area as having the highest risk of TB endemicity. Cluster 2 includes areas with moderate population density and TB case numbers, while Cluster 3 consists of subdistricts at higher altitudes with fewer TB cases. The clustering results were evaluated using three key metrics: Silhouette Score, Davies-Bouldin Index, and Dunn Index, which indicated that the clustering model performed well, although some subdistricts were positioned near the cluster boundaries. This research provides valuable information for health authorities to prioritize interventions and allocate resources to areas most in need of TB management.

A comparative analysis of unsupervised machine-learning methods in PSG-related phenotyping.

Obstructive sleep apnea is a heterogeneous sleep disorder with varying phenotypes. Several studies have already performed cluster analyses to discover various obstructive sleep apnea phenotypic clusters. However, the selection of the clustering method might affect the outputs. Consequently, it is unclear whether similar obstructive sleep apnea clusters can be reproduced using different clustering methods. In this study, we applied four well-known clustering methods: Agglomerative Hierarchical Clustering; K-means; Fuzzy C-means; and Gaussian Mixture Model to a population of 865 suspected obstructive sleep apnea patients. By creating five clusters with each method, we examined the effect of clustering methods on forming obstructive sleep apnea clusters and the differences in their physiological characteristics. We utilized a visualization technique to indicate the cluster formations, Cohen's kappa statistics to find the similarity and agreement between clustering methods, and performance evaluation to compare the clustering performance. As a result, two out of five clusters were distinctly different with all four methods, while three other clusters exhibited overlapping features across all methods. In terms of agreement, Fuzzy C-means and K-means had the strongest (κ = 0.87), and Agglomerative hierarchical clustering and Gaussian Mixture Model had the weakest agreement (κ = 0.51) between each other. The K-means showed the best clustering performance, followed by the Fuzzy C-means in most evaluation criteria. Moreover, Fuzzy C-means showed the greatest potential in handling overlapping clusters compared with other methods. In conclusion, we revealed a direct impact of clustering method selection on the formation and physiological characteristics of obstructive sleep apnea clusters. In addition, we highlighted the capability of soft clustering methods, particularly Fuzzy C-means, in the application of obstructive sleep apnea phenotyping.

FE-based machine learning model for predictive damage assessment in bonded composite joints via acoustic emission

This study systematically evaluated the mechanical performance of bonded composite single-lap joints with varying overlap lengths under tensile load. By integrating Acoustic Emission (AE) signals with Finite Element (FE) modeling, this study established a predictive machine learning model to enhance the understanding and prediction of failure mechanisms. Firstly, AE signal features were extracted and utilized as inputs for a hierarchical clustering model to classify AE signals into matrix cracking, fiber breakage, and adhesive debonding. AE signals classified as adhesive debonding were further analyzed by accumulating counts, amplitude, and energy. A cohesive zone-based Finite Element (FE) model was developed to simulate progressive debonding damage in the adhesive layer. After calibration with experimental results, various physical quantities such as stress, the damage initiation indicator, and the stiffness degradation indicator were extracted for correlation analysis. The experiments and simulations unravel the missing relationship between the AE and FE data. A strong correlation was observed between the cumulative count and cumulative amplitude with the damage initiation indicator, and between cumulative energy with the damage propagation indicator. Finally, based on the disclosed relationship between AE and FE data, a Support Vector Regression (SVR) model was established to predict the damage indicator. The developed model accurately predicted testing datasets within low absolute error ranges, underscoring its high predictive accuracy. Comprehensive stress analysis proposed a quantitative damage initiation indicator of 0.32 for structural fractures and a damage indicator of 0.12 for allowable stress levels. Consequently, combining AE techniques with FE modeling provided an effective tool for evaluating damage evolution in bonded composite joints.

A generative hierarchical topic model for image annotation

A redesigned hierarchical model for image annotation is presented in this study, which can also be simplified to image classification. Based on some previous models for textual analysis, such as LDA and the hierarchical clustering model by Hoffman et al. in 1988, the author applies similar ideas to image data, extracting image features as words and setting the latent variables of topics. Compared with the popular annotation methods which are deep-learning-based, the model put forward in this document details the generating process of images by adding the layer of topics. Besides, compared to other hierarchical models such as the multi-modal hierarchical model by Barnard and Forsyth in 2001, the model in this paper simplifies it to a single-modal one and simultaneously achieves similar effects. We collect images under different human activities, such as riding horses, playing the guitar, etc., mainly from Google search engine and the result proves that our model works better than most previous probabilistic models.

Art Therapy to Promote College Students’ Mental Health Based on a Hierarchical Clustering Algorithm

Art therapy is a therapeutic approach that utilizes the creative process of making art to improve mental, emotional, and physical well-being. Art therapy is a form of expressive therapy that utilizes the creative process of making art to improve mental, emotional, and psychological well-being. It provides individuals with a non-verbal outlet for self-expression and exploration, allowing them to communicate and process their thoughts, feelings, and experiences in a safe and supportive environment. This paper proposed an efficient Weighted Hierarchical Clustering Deep Neural Network (WH-CDNN) to promote the mental health of college students. The proposed WH-CDNN model extracts the features of the art therapy to promote the mental health of students. The features considered for the analysis are color palette, texture, and therapy for the promotion of mental health assessment of students. The features associated with the weighted model are computed for the college student mental health assessment. The features with the WH-CDNN model use the hierarchical clustering model for the computation of the features in art therapy based on the assessment of mental health. The examination is based on the consideration of 10 features for the estimation with the 5 clusters for the evaluation of the mental health assessment. Experimental analysis of the results demonstrated that the proposed WH-CDNN model achieves significant improvement in the after the art therapy of the students with the mental health assessment. Through simulation and analysis, the study demonstrates the effectiveness of art therapy in improving mental health outcomes, with significant reductions observed in anxiety and depression levels post-therapy. Moreover, the WH-CDNN model accurately predicts students' mental health states and evaluates the efficacy of art therapy interventions. The findings highlight the potential of integrating advanced computational techniques with art therapy to support student well-being and inform targeted mental health interventions in educational settings.

Differentiation between descending thoracic aortic diseases using machine learning and plasma proteomic signatures

BackgroundDescending thoracic aortic aneurysms and dissections can go undetected until severe and catastrophic, and few clinical indices exist to screen for aneurysms or predict risk of dissection.MethodsThis study generated a plasma proteomic dataset from 75 patients with descending type B dissection (Type B) and 62 patients with descending thoracic aortic aneurysm (DTAA). Standard statistical approaches were compared to supervised machine learning (ML) algorithms to distinguish Type B from DTAA cases. Quantitatively similar proteins were clustered based on linkage distance from hierarchical clustering and ML models were trained with uncorrelated protein lists across various linkage distances with hyperparameter optimization using fivefold cross validation. Permutation importance (PI) was used for ranking the most important predictor proteins of ML classification between disease states and the proteins among the top 10 PI protein groups were submitted for pathway analysis.ResultsOf the 1,549 peptides and 198 proteins used in this study, no peptides and only one protein, hemopexin (HPX), were significantly different at an adjusted p < 0.01 between Type B and DTAA cases. The highest performing model on the training set (Support Vector Classifier) and its corresponding linkage distance (0.5) were used for evaluation of the test set, yielding a precision-recall area under the curve of 0.7 to classify between Type B from DTAA cases. The five proteins with the highest PI scores were immunoglobulin heavy variable 6–1 (IGHV6-1), lecithin-cholesterol acyltransferase (LCAT), coagulation factor 12 (F12), HPX, and immunoglobulin heavy variable 4–4 (IGHV4-4). All proteins from the top 10 most important groups generated the following significantly enriched pathways in the plasma of Type B versus DTAA patients: complement activation, humoral immune response, and blood coagulation.ConclusionsWe conclude that ML may be useful in differentiating the plasma proteome of highly similar disease states that would otherwise not be distinguishable using statistics, and, in such cases, ML may enable prioritizing important proteins for model prediction.

Multiscale Urban Functional Zone Recognition Based on Landmark Semantic Constraints

The classification of urban functional areas is important for understanding the characteristics of urban areas and optimizing the utilization of urban land resources. Existing related methods have improved accuracy. However, they neglect cognitive differences amongst humans in the different scales of regional functions. Moreover, how to build the correlations of cross-scale characteristics is still unresolved when realizing the classification of multiscale urban functional zones. To resolve these problems, a transportation analysis zone involving urban buildings as research units is created and these units are described by geometric and functional characteristics using multiple data sources. Then, a hierarchical clustering model is built for the recognition of urban functional areas at varying scales with landmark semantic constraints. In the experiments, Shanghai served as the study area, and multiscale zones were created using different levels of road networks considering the constraint correlation of the significance between cross-scale maps. The experiential results show the proposed method has excellent performance and optimizes the functional zone classification at different scales. This study not only enriches the multiscale urban functional area-recognition methods but also can be used in other aspects, like cartographic generalization or spatial analysis.

Aviation Safety Risk Analysis Based on Bayesian Network Modeling

The occurrence of serious flight accidents not only brings huge economic losses to airlines but also poses a great threat to passengers' lives. The occurrence of serious flight accidents will not only bring great economic losses to airlines but also cause great threats to the lives of passengers. Therefore, in this paper, based on the flight records of different flight crews, flight routes, airports, and specific flight conditions, aircraft safety risk analysis is conducted to calculate and evaluate the risk propensity. Firstly, the data are preprocessed, and then the QAR data are clustered to extract the key data items affecting the safe flight of the aircraft; then the data are transformed to the attached data, and a hierarchical clustering model is established based on the K-means clustering algorithm, which classifies the factors affecting the safety quality of the aircraft flight into the three main aspects of environmental factors, crew maneuvering, and aircraft state, and the key parameters are weighted by the Random Forest algorithm. weights of the key parameters. Then, based on the 3 criterion, the outliers exceeding 3 were selected as the dependent variables leading to the occurrence of deviation; based on the three-layer model, a Bayesian neural network model based on the three-layer model, and then determine the Bayesian nodes; finally, calculate the Bayesian node probability, and then provide a reasonable quantitative description of the flight maneuver.

Mode-I analysis of glass fiber-reinforced polymer laminates toughened by interleaved polyamide 66 nanofibers and chopped carbon using acoustic emission

The fracture toughness of epoxy resins is weak, so enhancing its properties is essential. In this research it was found that interleaving chopped carbon fibers and polyamide 66 nano mats between fiberglass reinforced plastic laminates with 110 µm thickness, increased fracture toughness in Mode I by 35 and 112%, respectively. Acoustic emission technique showed that damage mechanisms begin at further displacements of load point in modified samples. A hierarchical clustering model was used to classify various failure types. It was observed that matrix cracking, fiber breakage, and matrix cracking were the major damage mechanisms for Reference, chopped carbon, and nanomodified samples, respectively. The cumulative energy released by nanomodified samples was 83% lower than that of Reference and chopped carbon samples. A scanning electron microscope was used to validate different types of damage mechanisms.

Prediction of groundwater contamination in an open landfill area using a novel hybrid clustering-based ai model

Groundwater is a vital resource that provides drinking water to over half of the world’s population. However, groundwater contamination has become a serious issue due to human activities such as industrialization, agriculture, and improper waste disposal. The impacts of groundwater contamination can be severe, including health risks, environmental damage, and economic losses. A list of unknown groundwater contamination sources has been developed for the Wang-Tien landfill using a groundwater modeling system (GMS). Further, AI-based models have been developed which accurately predict the contamination from the sources at this site. A serious complication with most previous studies using artificial neural networks (ANN) for contamination source identification has been the large size of the neural networks. We have designed the ANN models which use three different ways of presenting inputs that are categorized by hierarchical K-means clustering. Such an implementation reduces the overall complexity of the model along with high accuracy. The predictive capability of developed models was assessed using performance indices and compared with the ANN models. The results show that the hybrid model of hierarchical K-means clustering and ANN model (HCA-ANN) is a highly accurate model for identifying pollution sources in contaminated water.

Unveiling Longitudinal Patterns in International Mental Health Assessments: A Systematic Evaluation of Clustering Techniques

Background: Mental health assessments across diverse populations provide valuable insights into the prevalence and patterns of mental health issues. However, the complexity and volume of longitudinal data present challenges in extracting meaningful information for effective intervention. Clustering methods have emerged as powerful tools for identifying hidden structures within such datasets, yet a comprehensive evaluation of these techniques in the context of international mental health assessments is lacking. Objectives: This study aims to systematically evaluate various clustering techniques applied to longitudinal mental health data from international assessments. The focus is on understanding how different methods capture and reveal patterns and subgroups within the data, thereby guiding targeted mental health interventions. Methods: We applied and compared three clustering techniques—K-Means Clustering, Hierarchical Clustering, and Gaussian Mixture Models (GMM)—to longitudinal mental health assessment data. We assessed the performance of these methods in identifying meaningful clusters, considering their strengths and limitations in capturing the complexity of mental health trajectories. Results: Our analysis revealed distinct clusters reflecting varying levels of mental health severity and symptom trajectories. K-Means identified broad clusters, while Hierarchical Clustering provided insights into the data’s hierarchical structure. GMM offered a probabilistic view, highlighting overlapping mental health experiences among individuals. Each method contributed uniquely to understanding the longitudinal patterns in the data. Implications: The findings underscore the importance of using a multi-faceted approach to clustering in mental health research. By revealing different dimensions of mental health trajectories, this study provides valuable insights for tailoring interventions and resource allocation. The results highlight the need for ongoing evaluation of clustering techniques to enhance their applicability in diverse international contexts.

Monogenetic volcanoes as windows into transcrustal mush: A case study of Slamet and Loyang volcanoes, Central Java

Monogenetic volcanic fields (MVFs) are commonly associated with long-lived polygenetic composite volcanoes in many tectonic settings. The association between polygenetic and monogenetic volcanoes has raised questions as to the source of melts feeding monogenetic volcanoes – specifically, whether they are derived from the transcrustal mush underlying a nearby polygenetic volcano, or from a batch of parental magma derived from a different mantle source. To address this question, we studied a well-constrained suite of scoria samples from Gunung Slamet in Central Java, Indonesia, and its most distal monogenetic scoria cone, Gunung Loyang. Slamet is composed of calc-alkaline basalts, and has an extensive MVF on its eastern flank, similar to other Indonesian volcanoes. We analyzed the textural and chemical features of the olivine cargo in both volcanoes, including olivine- and clinopyroxene-hosted melt inclusions. These melt inclusions are the first such analyzed for volatile element contents (H2O, CO2) in the Slamet-Loyang system. We applied hierarchical clustering and mass balance models to constrain the features of the olivine populations in both systems. The olivines in both volcanoes cluster in five distinct textural groups, defined by the presence or lack of zoning and resorption, and according to the type and degree of zoning they exhibit. Olivines show a range of moderate to evolved Fo%, and they also show significant variation in nickel contents (50–1600 ppm). By estimating saturation pressures of entrapped melt inclusions, we observe that Loyang olivines are drawn from a similar petrological system as Slamet, but from deeper mid-crustal levels (12 km), and more mafic source. Our work suggests that monogenetic volcanoes like Loyang can be used as “petrological windows” into the deeper crustal levels of mush reservoirs at composite stratovolcanoes and their associated MVFs.

Machine Learning‐Aided Data Analysis in Single‐Protein Conductance Measurement with Electron Tunneling Probes†

Comprehensive SummaryThe electrical tunneling sensors have excellent potential in the next generation of single‐molecule measurement and sequencing technologies due to their high sensitivity and spatial resolution capabilities. Electrical tunneling signals that have been measured at a high sampling rate may provide detailed molecular information. Despite the extraordinarily large amount of data that has been gathered, it is still difficult to correlate signal transformations with molecular processes, which creates great obstacles for signal analysis. Machine learning is an effective tool for data analysis that is currently gaining more significance. It has demonstrated promising results when used to analyze data from single‐molecule electrical measurements. In order to extract meaningful information from raw measurement data, we have combined intelligent machine learning with tunneling electrical signals. For the purpose of analyzing tunneling electrical signals, we investigated the clustering approach, which is a classic algorithm in machine learning. A clustering model was built that combines the advantages of hierarchical clustering and Gaussian mixture model clustering. Additionally, customized statistical algorithms were designed. It has been proven to efficiently gather molecular information and enhance the effectiveness of data analysis.

Crop Information Retrieval Framework Based on LDW-Ontology and SNM-BERT Techniques

Currently, on the Internet, the information about agriculture is augmenting extremely; thus, searching for precise, relevant data of various details is highly complicated. To deal with particular difficulties like lower relevancy rate, false detection of retrieval resources, poor similarity rate, unstructured data format, multivariate data, irrelevant spelling, and higher computation time, an intelligent Information Retrieval (IR) system is required. An IR Framework centered on Levenshtein Distance Weight-centric Ontology (LDW-Ontology) and Sutskever Nesterov Momentum-centred Bidirectional Encoder Representation from Transformer (SNM-BERT) methodologies is presented here to overcome the complications as mentioned earlier. Firstly, the data is pre-processed, transmuting the unstructured data into a structured format, thus mitigating the error probabilities. Then, the LDW-Crop Ontology construction is done regarding the structured data. In the methodology presented, significance, frequency,and the suggestion of word in mind are considered to build Crop ontology. In the MongoDB database, the data being constructed are amassed. Then, by utilizing SNM-BERT, the data is trained for IR regarding clustered input produced by Inter Quartile Pruning Range-centred Hierarchical Divisive Clustering (IQPR-HDC) model. The LDW is computed for the provided user query; subsequently, the similarity evaluation outcomes are obtained from the database. The experiential evaluation displays that when analogized with the prevailing methodologies, a better accuracy of 94 % for simple queries and 92% for complex queries is achieved. Along with retrieval rate with lower computation time is achieved by the proposed methodology.

Explainable semi-supervised AI for green performance evaluation of airport buildings

Despite the global sustainability trend, airport buildings have received limited attention regarding their environmental impact within the built environment and aviation sectors. With less than 1% of green building certifications worldwide, there is a lack of essential green performance assessment tools, including frameworks, primary datasets, and models that require less human supervision. Previous research has addressed these issues by proposing frameworks, primary datasets, and a supervised method that leverages Classification and Regression Trees. However, these methods still rely on human effort to fix credit scores and assign category weights, necessitating significant supervision from human experts. This study presents a green performance evaluation model for airport buildings using an explainable semi-supervised AI approach to reduce human dependency during the inference of green scores. Principal Component Analysis and Hierarchical Clustering algorithms (agglomerative and divisive) are employed to construct the models. The dataset's dimensionality is reduced based on metrics such as scatter coefficient, Psi Index, variation, and permutation. The cluster numbers are determined using elbow point and silhouette coefficient values, resulting in cluster initializations between 6 and 12. Performance measures, including the rand index, mean absolute error, precision, recall, F1-measure, and accuracy, are evaluated for the models. The AHC and DHC models exhibit a maximum accuracy of 74%, albeit with different cluster numbers. Consequently, models are developed with suitable cluster numbers to derive green rules with improved predictive accuracy. Each agglomerative and divisive hierarchical clustering model yields 12 practical green rules that can assist airport operators and facility managers in enhancing the green performance of their airport buildings. These explainable green rules serve as guidelines to improve airport facilities. Additionally, a web-based green rating tool is developed to demonstrate the proof-of-concept, utilizing the green rules extracted from the agglomerative model.

Cultivation of Melilotus officinalis as a source of bioactive compounds in association with soil recovery practices.

Melilotus officinalis is a Leguminosae with relevant applications in medicine and soil recovery. This study reports the application of Melilotus officinalis plants in soil recovery and as a source of bioactive compounds. Plants were cultivated in semiarid soil under four different fertilizer treatments, urban waste compost at 10 t/ha and 20 t/ha, inorganic fertilizer and a control (no fertilizer). Agronomic properties of soil (pH, EC, soil respiration, C content, macro- and microelements) were analyzed before and after treatment. Also, germination, biomass, element contents, and physiological response were evaluated. Metabolite composition of plants was analyzed through Fourier-transform ion cyclotron resonance mass spectrometry (FT-ICR MS). Results showed a significant enhancement of the soil microbial activity in planted soils amended with compost, though there were no other clear effects on the soil physicochemical and chemical characteristics during the short experimental period. An improvement in M. officinalis germination and growth was observed in soils with compost amendment. Metabolite composition of plants was analyzed through Fourier-transform ion cyclotron resonance mass spectrometry (FT-ICR MS). Principal Component and Agglomerative Hierarchical Clustering models suggest that there is a clear separation of the metabolome of four groups of plants grown under different soil treatments. The five most important discriminative metabolites (annotated) were oleamide, palmitic acid, stearic acid, 3-hydroxy-cis-5-octenoylcarnitine, and 6-hydroxynon-7- enoylcarnitine. This study provides information on how the metabolome of Melilotus might be altered by fertilizer application in poor soil regions. These metabolome changes might have repercussions for the application of this plant in medicine and pharmacology. The results support the profitability of Melilotus officinalis cultivation for bioactive compounds production in association with soil recovery practices.

Customer Segmentation using Machine Learning Techniques

The objective of this customer segmentation endeavor is to utilize machine learning methodologies, Python programming, and the Streamlit framework in order to deliver individualized suggestions pertaining to savings plans, loans, and strategies for wealth management. Through the division of customers according to their demographic data, purchase patterns, and online navigation tendencies, this initiative empowers financial establishments to gain deeper insights into their customer demographics and customize their services to cater to specific requirements.&#x0D; The project involves several stages, including data collection, data preprocessing, feature selection, model training, and evaluation. The acquired customer data is refined to resolve any inconsistencies or missing information. Important characteristics are identified, and an exploratory analysis of the data is conducted to uncover hidden patterns and relationships. We then select an appropriate machine learning algorithm, such as K-means clustering, hierarchical clustering, or Gaussian Mixture Models, to categorize the customer base into distinct clusters.&#x0D; After the completion of model training and assessment, a Streamlit-based web application is constructed, offering an interactive platform for users. This application permits users to input their particulars and obtain tailored suggestions grounded in their designated segment. The suggestions encompass fitting savings schemes, loan alternatives, and wealth management approaches that harmonize with their financial objectives and risk appetite.&#x0D; The deployed application facilitates a seamless user experience, providing real-time recommendations and insights. Continuous improvement of the model and application is encouraged through user feedback, allowing for refinement and better customization of recommendations over time.

Autoantibody status according to multiparametric assay accurately estimates connective tissue disease classification and identifies clinically relevant disease clusters

ObjectiveAssessment of circulating autoantibodies represents one of the earliest diagnostic procedures in patients with suspected connective tissue disease (CTD), providing important information for disease diagnosis, identification and prediction of potential...

Semi-automated Hierarchical Clustering Model for 4D-STEM Datasets.

Semi-automated Hierarchical Clustering Model for 4D-STEM Datasets.

Analysis of glass artifacts based on hierarchical clustering and regression models

Glass is a valuable evidence of the early friendly trade between China and the Western countries. In this study, the rationality and sensitivity of the developed model were first evaluated by analyzing the definitive and quantitative data of each type of glass artifacts and establishing a model to classify the glass artifact categories into subcategories, as well as categorizing unknown types of glass. Considering the discrete and discontinuous nature of the data, the chi-square test was chosen to investigate the correlation between glass type, decoration and color and surface weathering. Secondly, descriptive statistics of high potassium and lead-barium glass artifacts before and after weathering were conducted to obtain statistical patterns about the chemical composition content, and then the chemical composition content of glass artifacts before weathering was predicted based on the mean rate of change of the chemical composition content of high potassium and lead-barium glass before and after weathering, respectively. Finally, this study further explored the relationship between glass artifacts and the content of each chemical composition using Spearman's correlation coefficient to derive the basis for the classification of glass types, based on the hierarchical clustering method to classify the subclasses of high potassium glass and lead-barium glass with and without weathering, respectively, and used the elbow rule to derive the most appropriate number of subclasses for the above four types of glass, which endowed the model with rationality.