Hierarchical Clustering Algorithm Research Articles

E-Calib: A Fast, Robust and Accurate Calibration Toolbox for Event Cameras.

Event cameras triggered a paradigm shift in the computer vision community delineated by their asynchronous nature, low latency, and high dynamic range. Calibration of event cameras is always essential to account for the sensor intrinsic parameters and for 3D perception. However, conventional image-based calibration techniques are not applicable due to the asynchronous, binary output of the sensor. The current standard for calibrating event cameras relies on either blinking patterns or event-based image reconstruction algorithms. These approaches are difficult to deploy in factory settings and are affected by noise and artifacts degrading the calibration performance. To bridge these limitations, we present E-Calib, a novel, fast, robust, and accurate calibration toolbox for event cameras utilizing the asymmetric circle grid, for its robustness to out-of-focus scenes. E-Calib introduces an efficient reweighted least squares (eRWLS) method for feature extraction of the calibration pattern circles with sub-pixel accuracy and robustness to noise. In addition, a modified hierarchical clustering algorithm is devised to detect the calibration grid apart from the background clutter. The proposed method is tested in a variety of rigorous experiments for different event camera models, on circle grids with different geometric properties, on varying calibration trajectories and speeds, and under challenging illumination conditions. The results show that our approach outperforms the state-of-the-art in detection success rate, reprojection error, and pose estimation accuracy.

Unsupervised Single-Cell Clustering with Asymmetric Within-Sample Transformation and Per-Cluster Supervised Features Selection.

This chapter shows applying the Asymmetric Within-Sample Transformation to single-cell RNA-Seq data matched with a previous dropout imputation. The asymmetric transformation is a special winsorization that flattens low-expressed intensities and preserves highly expressed gene levels. Before a standard hierarchical clustering algorithm, an intermediate step removes noninformative genes according to a threshold applied to a per-gene entropy estimate. Following the clustering, a time-intensive algorithm is shown to uncover the molecular features associated with each cluster. This step implements a resampling algorithm to generate a random baseline to measure up/downregulated significant genes. To this aim, we adopt a GLM model as implemented in DESeq2 package. We render the results in graphical mode. While the tools are standard heat maps, we introduce some data scaling to clarify the results' reliability.

DG Comics: Semi-Automatically Authoring Graph Comics for Dynamic Graphs.

Comics are an effective method for sequential data-driven storytelling, especially for dynamic graphs-graphs whose vertices and edges change over time. However, manually creating such comics is currently time-consuming, complex, and error-prone. In this paper, we propose DG COMICS, a novel comic authoring tool for dynamic graphs that allows users to semi-automatically build and annotate comics. The tool uses a newly developed hierarchical clustering algorithm to segment consecutive snapshots of dynamic graphs while preserving their chronological order. It also presents rich information on both individuals and communities extracted from dynamic graphs in multiple views, where users can explore dynamic graphs and choose what to tell in comics. For evaluation, we provide an example and report the results of a user study and an expert review.

Analysis of the changes in the art curriculum of preschool children’s education in the new media era

Abstract Art education for preschool children is the basis for the cultivation of young children’s perceptual ability, and the cultivation of young children’s perceptual ability plays an essential and fundamental role in young children’s understanding of the world, feeling of beauty, and discovery of beauty. This paper uses AR technology and microclassrooms in the context of new media to construct an educational model of preschool children’s art curriculum, and a portrait of preschool children’s scholars is built. It focuses on combining the hierarchical clustering algorithm with the radial function method of multivariate interpolation to analyze the classroom behaviors of preschool children, mining the correlation between the course learning behaviors and the learning effect through the Apriori algorithm, and constructing a personalized art learning path for preschool children based on the ant colony algorithm for practice. The results showed that preschool children were divided into four categories, with the highest proportion being excellent preschool children (52.5%). The mean score of preschool children’s art perception score was 8.225, and their scores were normally distributed, with the highest number of preschool children between 8.00 and 8.25. This study is significant for changing the art curriculum in preschool children’s education. Teachers should be guided in teaching activities and focus on children’s perceptual experiences and expressions.

Radio Frequency Identification Network Planning Using Ascendant Hierarchical Clustering Algorithm

Radio Frequency Identification (RFID) is a suitable technology for several application scenarios. However, it faces many issues that need to be addressed. RFID Network Planning (RNP) is one of these challenges, which is studied in this paper. It consists in defining the optimal placement of RFID readers and their communication range. In this study, the RNP issue is considered as a clustering problem. Therefore, it is solved using Ascendant Hierarchical Clustering algorithm (AHC) such that each cluster head represents a reader. The advantage of our proposed approach is that, in addition to optimizing the RNP, it is capable to determine the optimal number of readers needed to get optimal performances. Simulation results demonstrate that the proposed AHC algorithm performs better than two efficient RNP algorithms from the literature: Particle Swarm Optimization (PSO) and Leaders and Followers (LaF) algorithms. Hence, it is observed that AHC algorithm is more efficient, scalable and runs faster than PSO and LaF.

Algorithm Design and Application of Grammatical Structure and Function Learning in Chinese Writing for Intelligent Analysis

Abstract This study utilizes the Larsen-Freeman three-dimensional theory of grammar teaching to quantitatively analyze the grammatical attributes “form”, “meaning” and “usage” in the Chinese writing process. The study uses the PCFG model to extract the syntactic dependency tree and represents the features of the tree structure with the help of TF-IDF. Meanwhile, the Transformer model extracts the semantic and syntactic features of sentences. At the same time, the positional encoder is removed to ensure that the model obtains information from the syntactic level. Further, an unsupervised method for generating grammatical textual recapitulation for Chinese Writing is proposed, as well as a Grammar Tree Generator for generating recapitulated sentences, which efficiently extracts the features of the input grammatical sequences. In addition, the study also includes writing feature analysis based on grammatical function matching, using a hierarchical clustering algorithm to analyze the similarity of 60 grammatical functions. Finally, validation was performed on 30 Chinese writing text collections, each containing 10 compositions, and the results showed high accuracy of unlabeled grammatical function recognition. The LDA model determined the optimal number of writing topics to be 150. This study highlights the potential application of intelligent analysis techniques in improving the quality of Chinese Writing. It provides new perspectives for an in-depth understanding of the interplay between grammatical structures and functions in the Chinese writing.

A Recognition Model Incorporating Geometric Relationships of Ship Components

Ship recognition with optical remote sensing images is currently widely used in fishery management, ship traffic surveillance, and maritime warfare. However, it currently faces two major challenges: recognizing rotated targets and achieving fine-grained recognition. To address these challenges, this paper presents a new model called Related-YOLO. This model utilizes the mechanisms of relational attention to stress positional relationships between the components of a ship, extracting key features more accurately. Furthermore, it introduces a hierarchical clustering algorithm to implement adaptive anchor boxes. To tackle the issue of detecting multiple targets at different scales, a small target detection head is added. Additionally, the model employs deformable convolution to extract the features of targets with diverse shapes. To evaluate the performance of the proposed model, a new dataset named FGWC-18 is established, specifically designed for fine-grained warship recognition. Experimental results demonstrate the excellent performance of the model on this dataset and two other public datasets, namely FGSC-23 and FGSCR-42. In summary, our model offers a new route to solve the challenging issues of detecting rotating targets and fine-grained recognition with remote sensing images, which provides a reliable foundation for the application of remote sensing images in a wide range of fields.

Robust prediction for CH4/CO2 competitive adsorption by genetic algorithm pruned neural network

Capturing the competitive adsorption behavior of CH4/CO2 in shale reservoirs is essential for estimating the original gas-in-place and enhancing the shale gas recovery. This study leverages the neural network with sparsity, trained on a comprehensive dataset that includes crucial parameters such as clay, total organic carbon, specific surface area, pressure, and initial composition. Sparsity diverges from Dropout by selectively pruning unnecessary connections to enhance the model's robustness. Specifically, previous attempts suffer from over-optimistic performance due to unsuitable random dataset sampling. Hierarchical clustering and permutation importance algorithm indicate that petrophysical properties including clay, total organic carbon and specific surface area. By optimizing the sparsity through a genetic algorithm and connection pruning strategy, an accuracy improvement from 0.90 to 0.95 in terms of R-squared compared to conventional fully connected networks is witnessed. Additionally, the results show that removing up to 40% of the connections without compromising accuracy. Therefore, a heterogeneous configuration of activation functions within the neural network further amplifies its predictive capability. In addition, compared to the Dropout strategy, sparsity is more advantageous for generalizability in scenarios with limited datasets.

Mineralogical and elemental composition of the Middle Miocene coal seams from the Alpu coalfield (Eskişehir, Central Türkiye): Insights from syngenetic zeolite formation

This study focuses on determining mineralogical and elemental compositions of coal seams (to the upwards, D, C, B, A, and S0) within seven coal exploration wells from the Alpu coalfield (Eskişehir, Central Türkiye). Furthermore, the special goal of the study is a comparative analysis of the relations between the elements by using agglomerative hierarchical clustering algorithm with different linkage methods as well as different similarity measures. Clay minerals and quartz are commonly detected as abundant to dominant phases, while natural zeolite formations were detected in the studied seams C, B, A, and S0. The SEM-EDX data shows that clinoptilolites in zeolite minerals were observed within the organic matter, while crystalline and non-crystalline analcime minerals along with syngenetic authigenic rhomboid K-feldspars were only detected in the seam A from one studied well. The existence of some micron-sized minerals such as apatite, monazite, and Ti-oxides within the smectite matrix and the measurable amount of Ti in smectite imply that alteration of epiclastic and contemporaneous volcanic inputs was developed under weak acidic to neutral conditions during peat accumulation. The lack of natural zeolite and carbonate minerals in the seam D could be an indicator of weak acidic to neutral conditions and semi-closed hydrogeological conditions. Nevertheless, the alkalinity of mire water, water table, and hydrogeological regime seem to be variable during the accumulation of precursor peats of seams C, B, A, and S0. In turn, alteration of volcanic inputs was observed under neutral to weak alkaline conditions and semi-closed to closed hydrogeological regime. Hence, syngenetic authigenic, micron-sized clinoptilolites were formed. Moreover, the existence of authigenic rhomboid K-feldspars and syngenetic authigenic analcimes in certain exploration well could suggest local increases on dissolved Na+ concentrations, alkalinity, and water table. Except for volcanogenic origin for minerals, accessory micron-sized minerals, like chromite, pentlandite, and allanite grains presumably originated from clastic influxes of ophiolitic rocks in the basement into palaeomires. The variations in mire water chemistry and clastic-influx source area could also control the elemental enrichments in the studied seams. Epiclastic and contemporaneous volcanic inputs into palaeomires seem to control enrichments of Li, B, Sc, and Ti in coal samples, while clastic influx from ophiolitic rocks into palaeomires caused to enrichments of Cr, V and Ni. Furthermore, the liberated Ba, Sr, and As ions from the alteration of epiclastic and contemporaneous volcanic inputs are absorbed by syngenetic zeolite minerals, while anoxic conditions in the palaeomires resulted in precipitation of Sr-barite and As-bearing pyrite grains during peat accumulation and/or early diagenetic stages. Overall, the differences in water chemistry of mire water, epiclastic and contemporaneous volcanic inputs, and clastic influx from the adjacent areas also caused several elemental enrichments and variations in mineralogical compositions of the Middle Miocene coal seams in the Alpu coalfield.

Trait‐based analyses reveal global patterns in diverse diets of albacore tuna (Thunnus alalunga)

AbstractSimplifying complex species interactions can facilitate tracking and predicting functional responses to ecological stressors. This is important for highly migratory pelagic predators, exploiting diverse prey fields as they respond to dynamic environments. We reconstructed the historical resource use of albacore tuna (Thunnus alalunga) globally from the 1900s to 2015 and confirmed highly biodiverse diets with 308 prey species, and an additional 238 taxa at coarser taxonomic resolution. We synthesised prey diversity into seven functional trait guilds using hierarchical divisive clustering algorithms as a function of four traits that describe habitat use and influence predator–prey encounter rates – prey habitat association vertically in the water column, horizontally along the coastal to pelagic gradient, seasonal and diel vertical migratory behaviour. We explored variability in historical composition of albacore diets across geographies based on species identity, individual trait information and functional trait guilds using a multi‐matrix modelling framework. Taxonomic information remains important for trophic ecology, however, species‐based diet composition in albacore tuna was highly variable across geographies and years sampled, making interpretation of these patterns difficult. By simplifying species identity into habitat‐based traits, we highlight changes in prey resources consumed, such as the historical importance of near‐surface epipelagic prey resources from coastal to oceanic habitats, and seasonally migrating continental shelf prey, with less frequent pulses of deeper water and demersal taxa. Trait information and trait guilds serve as useful classification frameworks for identifying functionally redundant food web linkages across biodiverse prey, and will prove useful in tracking predators' foraging responses to changing resource availability.

Risk-based regionalization approach for area-wide management of HLB vectors in the Mediterranean Basin.

Huanglongbing (HLB) is one of the most devastating citrus diseases worldwide. It is associated with the non-culture bacteria Candidatus Liberibacter spp., which can be transmitted by grafting and/or the psyllid vectors Diaphorina citri (ACP) and Trioza erytreae (AfCP). Although HLB has not been reported in the Mediterranean Basin to date, both vectors are present, and thus represent a serious threat to the citrus industry in this region. Resistant citrus cultivars or effective therapeutic treatments are not currently available for HLB. Nevertheless, area-wide pest management via coordinated management efforts over large areas has been implemented in Brazil, China and the USA for HLB control. This study proposes an open access flexible methodology to address area-wide management of both HLB vectors in the Mediterranean Basin. Based on a risk-based approach which considers climatic information and other variables that may influence vector introduction and spread, such as conventional, organic, abandoned and residential citrus areas as well as transportation corridors, an area-wide management division in pest management areas (PMAs) is proposed. The size and location of these PMAs were estimated by means of a hierarchical clustering algorithm with spatial constraints whose performance was assessed under different configuration scenarios. This proposal may assist policymakers and the citrus industry of the citrus-growing areas of the Mediterranean Basin in risk management planning in the case of the spread of HLB vectors or a possible introduction of the disease. Additionally, it may be a valuable resource to inform opinion dynamic models, enabling the identification of pivotal factors for the success of control measures.

Clustering of students admission data using k-means, hierarchical, and DBSCAN algorithms

Admissions in the university undergo procedures and requirements before a student can be officially enrolled. The senior high school grades remain the most significant in college admission decisions. This paper presents the use of data mining to cluster students based on admission datasets. The admission dataset for 2019-2020 was obtained from the office of student affairs and services. This dataset contains 2,114 observations with 11 attributes. Data preparation and data standardization were performed to ensure that the dataset is ready for processing and implemented in R programming language. The optimal number of clusters (k) was identified using the silhouette method. This method gave an optimal number of k=2 which was used in the actual clustering using the k-means and hierarchical clustering algorithms. Both algorithms were able to cluster students into two: cluster 1-social sciences or board courses and cluster 2-management or non-board courses. Further, density-based spatial clustering of applications with noise (DBSCAN) clustering algorithm was also used on the same dataset and it yielded a single cluster. This study can be replicated by using at least a 5-year dataset of students’ admission data employing other algorithms that would suggest students’ retention and turn over to board examinations.

Distribution Network Optimization Dispatch Method Based on Renewable Energy Scenario Reduction

The scenario analysis method is an important method to adapt to the optimal dispatch of power systems with a high percentage of new energy sources. As a hot research topic in scenario analysis methods, the significance of scenario reduction is to describe many complex scenarios that feature a small number of representative scenarios to achieve the purpose of reducing computational complexity. In this paper, a distribution network optimization dispatch method based on the hierarchical clustering algorithm for scenario reduction is proposed considering wind power output and photovoltaic output. Firstly, the original scenes are quickly categorized because of the hierarchical clustering algorithm to derive typical scenarios. Secondly, an optimal dispatch model of the distribution network containing renewable energy is established. Finally, the arithmetic example is verified by using the IEEE33 distribution network, and the experimental results validate the validity and superiority of the proposed scenario reduction method.

Opinion evolution and dynamic trust-driven consensus model in large-scale group decision-making under incomplete information

The shift to a new era of dealing with big data has driven continuous progress and development in computer science, artificial intelligence and machine learning. This change has led to the application of advanced techniques in the realm of decision science, particularly in the area of large-scale group decision-making (LSGDM). However, although these existing techniques have become the core of LSGDM methods, they are still limited in solving problems facing incomplete data. In addition, due to the rise of social media platforms such as Weibo, WeChat and Twitter, which build bridges for communication between decision makers (DMs), this brings new opportunities and challenges for consensus research. To address this set of issues, this study develops a consensus architecture that combines dynamic social network and opinion evolution in the context of an incomplete multi-attribute LSGDM. It is worth mentioning that the proposed consensus framework is a novel decision-making system that can be used to complete the estimation of the missing values and the consensus reaching process (CRP) by simulating the realistic decision-making scenarios. Firstly, considering the size of the trust value and the length of the path, a new trust propagation method is designed to achieve a more reliable estimation of the unknown trust value. Secondly, this paper establishes a missing value estimation method by virtue of the improved DeGroot model, which is able to obtain complete evaluation information by simulating the opinion formation process of DMs. Next, a hierarchical clustering algorithm with stronger robustness is constructed, which not only can adaptively complete the clustering process, but also integrally considers two attributes of trust and opinion similarity. In light of the above research, this study designs an opinion evolution and dynamic trust-driven consensus model, referred to as the DSN-DG-LSGDM model. Finally, the sensitivity analysis and experiments on a real dataset verify the significant superiority of the constructed DSN-DG-LSGDM model compared with the extant LSGDM consensus models.

How Do Heterogeneous Networks Affect a Firm's Innovation Performance? A Research Analysis Based on Clustering and Classification.

Based on authorized patents of China's artificial intelligence industry from 2013 to 2022, this paper constructs an Industry-University-Research institution (IUR) collaboration network and an Inter-Firm (IF) collaboration network and used the entropy weight method to take both the quantity and quality of patents into account to calculate the innovation performance of firms. Through the hierarchical clustering algorithm and classification and regression trees (CART) algorithm, in-depth analysis has been conducted on the intricate non-linear influence mechanisms between multiple variables and a firm's innovation performance. The findings indicate the following: (1) Based on the network centrality (NC), structural hole (SH), collaboration breadth (CB), and collaboration depth (CD) of both IUR and IF collaboration networks, two types of focal firms are identified. (2) For different types of focal firms, the combinations of network characteristics affecting their innovation performance are various. (3) In the IUR collaboration network, focal firms with a wide range of heterogeneous collaborative partners can obtain high innovation performance. However, focal firms in the IF collaboration network can achieve the same aim by maintaining deep collaboration with other focal firms. This paper not only helps firms make scientific decisions for development but also provides valuable suggestions for government policymakers.

TeraHAC: Hierarchical Agglomerative Clustering of Trillion-Edge Graphs

We introduce TeraHAC, a (1+ε)-approximate hierarchical agglomerative clustering (HAC) algorithm which scales to trillion-edge graphs. Our algorithm is based on a new approach to computing (1+ε)-approximate HAC, which is a novel combination of the nearest-neighbor chain algorithm and the notion of (1+ε)-approximate HAC. Our approach allows us to partition the graph among multiple machines and make significant progress in computing the clustering within each partition before any communication with other partitions is needed. We evaluate TeraHAC on a number of real-world and synthetic graphs of up to 8 trillion edges. We show that TeraHAC requires over 100x fewer rounds compared to previously known approaches for computing HAC. It is up to 8.3x faster than SCC, the state-of-the-art distributed algorithm for hierarchical clustering, while achieving 1.16x higher quality. In fact, TeraHAC essentially retains the quality of the celebrated HAC algorithm while significantly improving the running time.

Artificial intelligence-driven echocardiographic classification identifies Brugada Syndrome patients at higher arrhythmic risk

Abstract Background Mounting evidence exists to support the hypothesis that patients with Brugada Syndrome (BrS) present morphological and structural anomalies localized to the right ventricle (RV), but this has not been investigated in a large cohort and the association with clinical outcomes are unknown. Purpose To investigate the role of echocardiography-derived RV parameters in a large, single-center cohort of patients with BrS. Methods We prospectively enrolled n=492 consecutive patients (379 males, 77%; age 46±14 years) with confirmed BrS who underwent an echocardiogram during their last visit at the outpatient clinic for inherited arrhythmogenic disorders. Comprehensive 2D and color Doppler echocardiographic evaluation was performed using the standard clinical protocol. Unsupervised clustering analysis using cluster package sorted body surface area-corrected morphological (RVOT linear dimensions and RV diastolic area) and functional measurements of RV (fractional area shortening [FAC], tricuspid annular plane systolic excursion [TAPSE]), blinded to other variables. We performed clustering with a hierarchical cluster algorithm (h-cluster) and the optimal number of clusters (k=2) was determined using fviz_nbclust-derived plots. Results All n=492 patients were classified into 2 stable clusters () with the best robustness. The Cluster 1 echocardiograms exhibited normal RV morphological and functional parameters. The Cluster 2 echocardiograms showed a dilatation of the RVOT (PLAX 17.2±2.1 vs. 15.6±2.1, p<0.001; PSAX 18.3±2.2 vs. 16.9±2.2, p<0.001) compared to Cluster 1, while the RV diastolic area was not different (9.9±1.5 vs. 9.9±1.6, p=0.801). Importantly, RV function as assessed by TAPSE was not different between the two groups, but Cluster 2 patients showed a significantly impaired radial contraction (RV FAC 40.1±3.4% vs. 48.6±3.8%, p<0.001). We then investigated the clinical characteristics of patients based on clusters. Patients in Cluster 1 were younger (43.2±12.8 years vs. 47.5±14.0 years, p<0.001) and had significantly less arrhythmic symptoms such as syncope (13% vs. 20%, p=0.046) and life-threatening arrhythmic events (2% vs. 7%, p=0.017). Importantly, no significant differences in terms of gender (77% males in both groups, p=1.000), pattern type (43% with spontaneous type 1 pattern in Cluster 1 vs. 47%, p=0.494) nor genetic status were observed. Multivariate logistic regression confirmed that at parity of age and arrhythmic syncope, Cluster 2 was independently associated with an increased incidence of life-threatening arrhythmic events (odds ratio 3.71, 95% confidence interval 1.31-10.53, p=0.039). Conclusions We constructed a novel, echocardiography-guided classification of patients with BrS using unsupervised clustering analysis. Our data suggest that RV structural abnormalities are associated with a higher incidence of life-threatening arrhythmic events in patients with BrS, independently of arrhythmic syncope or age.

Identification of circulating miRNA as early diagnostic molecular markers in malignant glioblastoma base on decision tree joint scoring algorithm.

The lack of clinical markers prevents early diagnosis of glioblastoma (GBM). Many studies have found that circulating microRNAs (miRNAs) can be used as early diagnostic markers of malignant tumours. Therefore, the identification of novel circulating miRNA biomolecular markers could be beneficial to clinicians in the early diagnosis of GBM. We developed a decision tree joint scoring algorithm (DTSA), systematically integrating significance analysis of microarray (SAM), Pearson hierarchical clustering, T test, Decision tree and Entropy weight score algorithm, to screen out circulating miRNA molecular markers with high sensitivity and accuracy for early diagnosis of GBM. DTSA was developed and applied for GBM datasets and three circulating miRNA molecular markers were identified, namely, hsa-miR-2278, hsa-miR-555 and hsa-miR-892b. We have found that hsa-miR-2278 and hsa-miR-892b regulate the GBM pathway through target genes, promoting the development of GBM and affecting the survival of patients. DTSA has better classification effect in all data sets than other classification algorithms, and identified miRNAs are better than existing markers of GBM. These results suggest that DTSA can effectively identify circulating miRNA, thus contributing to the early diagnosis and personalised treatment of GBM.

Abstract 18518: Exploring Sleep's Impact On Sex And Racial/ Ethnic Disparities In Cardiovascular Disease Risk: Leveraging Machine Learning And Precision Insights To Unravel Sleep-cvd Connections In 10-year Sleep Health Health Study

Introduction: The study aimed to explore the connection between insufficient sleep (<7 hrs.) and 10-year cardiovascular disease (CVD) risk levels in black and white individuals, using Framingham risk (FRS) and Reynolds risk (RRS) scores. It also sought to identify and characterize the CVD risk factors associated with heightened or reduced risk profiles. Methods: Using the Sleep Heart Health Study (SHHS) (N=6,441) data, we investigated the independent role of insufficient sleep in explaining differences in 10-years CVD between blacks and whites via a proportional odds model of four 10-year CVD risk groups: low (<5%), low-medium (5% to <10%), medium-high (10% to <20%) and high (≥20%), adjusting for age, sex, and apnea-hypopnea index (AHI). We performed two levels of cluster analyses; via hierarchical cluster algorithm with entire sample (Level 1), and latent profiles in the low (protective profiles) and high (risk profiles) CVD risk groups (Level 2) to determine overall CVD risk, and risk and protective CVD profiles. Results: Blacks had a higher prevalence of smoking behavior, diabetes, mean systolic blood pressure, body mass index, total cholesterol compared to whites. Conversely, whites had a higher mean HDL cholesterol, sleep hours, and sleep efficiency compared to blacks. Men had higher 10-year CVD risk than women. AHI and race/ethnicity-sleep interaction were positively associated, while sleep was negatively associated with FRS and RRS. Across all CVD risk groups, whites who slept less than 5.5 hrs. had a higher CVD risk and those who slept more than 6.5 hrs. had a lower CVD risk compared to blacks. In Level 1 cluster analyses, we found two clusters: Cluster 1 (n= 3233): 6.17 sleep hours, apnea-index 11.84, age 59, SBP 125.43, total cholesterol 209, HDL 51.39, BMI 29.03, and slightly more than 50% female; and Cluster 2 (n=1657): 5.61 sleep hours, apnea-index 13.41, age 74, SBP 131, total cholesterol 204, HDL 50.30, BMI 26.45, and slightly less than 50% female. In Level 2 cluster analyses, we found two profiles within the low and high CVD risk groups. Conclusions: These findings suggest that blacks may not receive full protection from long-term CVD risk with longer sleep duration, as their white counterparts. Support: R01HL142066, R01AG067523, R01AG075007

Boosting cluster tree with reciprocal nearest neighbors scoring

Clustering plays a pivotal role in knowledge processing, knowledge bases, and expert systems, enabling AI systems to acquire knowledge effectively. Hierarchical clustering, in particular, offers an intelligent approach to represent knowledge hierarchically by transforming raw data into one/multiple tree-shaped components. However, a notable difficulty arises when attempting to pinpoint appropriate representative points within lower levels of the cluster tree. These points are of paramount importance, as they serve as the roots for subsequent aggregation within the upper levels of the cluster tree. Traditional hierarchical clustering algorithms have relied on rudimentary techniques to select these representative points, which may not provide an adequate representation. Consequently, the resulting cluster tree often falls short in terms of empirical performance. To address this shortcoming, we proposed an innovative hierarchical clustering algorithm in this paper. The proposed algorithm is designed to efficiently identify the representative point within each sub-minimum-spanning-tree during the construction of the cluster tree, achieved by topology-based scoring the reciprocal nearest data points. Rigorous testing on UCI datasets has demonstrated the superior clustering accuracy (measured by Rand Index and Normalized Mutual Information) of our proposed algorithm compared to other benchmark algorithms. Further analysis reveals that our algorithm boasts a O(nlogn) time-complexity and a O(logn) space-complexity, indicating its scalability and efficiency in handling large-scale data with minimal time and storage costs. Importantly, our algorithm’s ability to process up to two million data points on a standard personal computer underscores its cost-effectiveness.