Agglomerative Approach Research Articles

Agglomerative approach to the analysis of wind regime over Lake Baikal during the existence of the autumnal thermal bar

An approach based on agglomerative hierarchical clustering of wind direction data is proposed to analyze the wind regime over the middle Baikal region. Wind direction data were obtained from an objective National Centers for Environmental Prediction (NCEP) analysis. The constructed maps agree well with the measurement data collected at the meteorological stations in the Irkutsk and Zabaykalsky departments of Hydrometeorological and Environmental Monitoring.

On the Effects of Similarity in Community Detection

Community detection in social networks is a significant area of research within Artificial Intelligence and social network analysis. The agglomerative method is a well-known approach used for detecting communities. This technique relies on local similarities, to form clusters between pairs of nodes in the graph. The process involves merging each pair of nodes that exhibits the highest similarity and then computing a quality function for the current clustering results; each such merge, along with the computation of the quality of results, constitutes a step or called iteration. The choice of a similarity function can play a crucial role in determining the number of iterations, which in turn affects the running time of an agglomerative method. This raises a fundamental question when employing an agglomerative approach: a worth-asking question is how to determine which similarity function to choose and why. To address this question, our paper delves into the comparison of two well-known similarity functions: structural similarity and hub-promoted similarity. We conducted a deep comprehensive theoretical analysis followed by an extensive experimentation on several datasets, focusing on the computational aspects of these functions. Notably, our findings highlight that in specific cases within the graph, the hub-promoted similarity function is faster compared to structural similarity.

One shot alpha numeric weight based clustering algorithm with user threshold

<p>Information Retrieval from Files and data bases like data sources is a major issue now days. After Information Retrieval clustering is also a one of the important things. In the market so many clustering algorithms were available. But choosing of the clustering algorithm depends on the user requirements. This paper addresses the study of agglomerative approach for different constraints or metrics or user preferences like Number of levels in the clustering process, number of clusters that should be generated at each level and range of the attributes at each level for doing the clustering for the given data set. In brief overview we discuss the agglomerative approach for clustering algorithm with their user preferences.</p>

Parallel Transformation of Bounding Volume Hierarchies into Oriented Bounding Box Trees

AbstractOriented bounding box (OBB) hierarchies can be used instead of hierarchies based on axis‐aligned bounding boxes (AABB), providing tighter fitting to the underlying geometric structures and resulting in improved interference tests, such as ray‐geometry intersections. In this paper, we present a method for the fast, parallel transformation of an existing bounding volume hierarchy (BVH), based on AABBs, into a hierarchy based on oriented bounding boxes. To this end, we parallelise a high‐quality OBB extraction algorithm from the literature to operate as a standalone OBB estimator and further extend it to efficiently build an OBB hierarchy in a bottom up manner. This agglomerative approach allows for fast parallel execution and the formation of arbitrary, high‐quality OBBs in bounding volume hierarchies. The method is fully implemented on the GPU and extensively evaluated with ray intersections.

A hierarchical cluster analysis of the psycological impact of the COVID-19 pandemic on Italian severe asthma patients

Introduction In the context of COVID-19 pandemic, a consistent medical concern raised among severe asthma patients, though the studies excluded an increased risk of severe disease as well as an increased susceptibility. The aim of the study was to apply the Psychological General Well-Being Index (PGWBI) questionnaire to severe asthmatics during the COVID-19 pandemic and to evaluate the data with a hierarchical cluster analysis. Methods 114 severe asthmatics were asked to respond anonymously to the PGWBI questionnaire. The patients underwent a lung functional test, fractional exhaled nitric oxide (FeNO) measurement, Asthma Control Test (ACT), and Asthma Control Questionnaire (ACQ6). A hierarchical cluster analysis was performed using an agglomerative approach and complete linkage to evaluate the results. Results The study population predominantly included female (60%), middle-aged patients, with normal lung function parameters, mild signs of airway, and satisfactory asthma control. The PGWBI score (82.46 ± 16.53) of the study population showed a good state of psychological well-being and was similar to that of a representative sample of healthy adult Italian subjects. Thus, Hierarchical cluster analysis identified 3 groups of patients: Cluster 1 (32%), Cluster 2 (64%), and Cluster 3 (4%). Whilst the Cluster 2 patients’ PGWBI score fell within the normal range, the Cluster 1 patients had a significantly lower total score (68.57 ± 7.2; p < 0.05), suggesting moderate distress. The Cluster 3 patients presented a total score markedly low. Conclusion Although the majority of the severe asthma patients studied demonstrated good mental well-being during the COVID-19 pandemic, some did indeed show moderate to severe psychological distress.

A systems level approach to study metabolic networks in prokaryotes with the aromatic amino acid biosynthesis pathway.

Metabolism of an organism underlies its phenotype, which depends on many factors, such as the genetic makeup, habitat, and stresses to which it is exposed. This is particularly important for the prokaryotes, which undergo significant vertical and horizontal gene transfers. In this study we have used the energy-intensive Aromatic Amino Acid (Tryptophan, Tyrosine and Phenylalanine, TTP) biosynthesis pathway, in a large number of prokaryotes, as a model system to query the different levels of organization of metabolism in the whole intracellular biochemical network, and to understand how perturbations, such as mutations, affects the metabolic flux through the pathway - in isolation and in the context of other pathways connected to it. Using an agglomerative approach involving complex network analysis and Flux Balance Analyses (FBA), of the Tryptophan, Tyrosine and Phenylalanine and other pathways connected to it, we identify several novel results. Using the reaction network analysis and Flux Balance Analyses of the Tryptophan, Tyrosine and Phenylalanine and the genome-scale reconstructed metabolic pathways, many common hubs between the connected networks and the whole genome network are identified. The results show that the connected pathway network can act as a proxy for the whole genome network in Prokaryotes. This systems level analysis also points towards designing functional smaller synthetic pathways based on the reaction network and Flux Balance Analyses analysis.

IMPULSE OSCILLOMETRY IDENTIFIES DISTINCT ASTHMA PHENOTYPES ON CLUSTER ANALYSIS

IMPULSE OSCILLOMETRY IDENTIFIES DISTINCT ASTHMA PHENOTYPES ON CLUSTER ANALYSIS

A regionalization method filtering out small-scale spatial fluctuations

Abstract. Regionalization is the process of aggregating contiguous spatial units to form areas that are homogeneous with respect to one or a set of variables. It is useful when studying spatial phenomena or when designing region-based policies, as it allows to unravel the latent spatial structure of a dataset. However, this task is challenging when small-scale fluctuations in the data interfere with the phenomenon of interest. In such circumstances, regionalization techniques are prone to overfitting small-scale fluctuations, and producing erratic regions. This paper presents a regionalization method robust to small-scale variations that is particularly relevant when handling demographic data. Fluctuations are filtered out using a weighted spatial average before applying agglomerative clustering. The method is tested against a conventional agglomerative clustering approach on a fine-resolution demographic dataset, for a set of indicators quantifying: the ability to identify large-scale spatial patterns, the homogeneity of the regions produced, and the spatial regularity of these regions. These indicators have been computed for the two methods for a number of clusters ranging from 2 to 101, and results show that the proposed approach performs better than conventional agglomerative clustering more than 90% of the time at identifying large-scale patterns, and produces more regular regions 96% of the time.

Agglomerative Approach to Assessing the Degree of Criminalization of the Subjects of the Russian Federation: the Methodology of Statistical Analysis and Criminological Description

Alongside a wide range of analytical crime research methods in criminology, the method of assessing the degree of criminalization of the subjects of the Russian Federation through an agglomerative approach to the features characterizing the indicators of regional crime is of special research interest. This method includes three stages: 1) sampling and processing empirical statistical indicators of regional crime; 2) clustering of regions by groups of criminological characteristics; 3) criminological analysis and summary of obtained results. In order to assess the criminalization of subjects and Districts of the Russian Federation, the authors use such indicators as the categories of crimes, as they are most informative ones regarding the degree of crime prevalence in the population. The calculations are based on crimes registered by category per 100 thousand residents in the corresponding subject or District. The authors use three classes of criminal activity of population to present the results of clustering which reflects the minimal number of elements necessary for the discrete covering of all the variety of manifestations of criminal social prevalence and makes it possible to characterize groups of regions with high, medium and low levels of such prevalence. Calculation results allowed the authors to interpret classes as groups of subjects with a low (first class), medium (second class) and high (third class) level of criminalization of their population. The authors also present the clustering of subjects and Federal Districts of the Russian Federation by coefficients of registered crimes of the corresponding categories. This clustering encompasses information in 2010–2020. The dynamics of the distribution of the subjects of the Russian Federation by class showed a wave-like transition of the specific weight of the corresponding classes and a stable growth of criminal prevalence in the population in 2019–2020. The scattering diagrams make it possible to use a trendline with linear parameters to show either a positive or a negative correlation of crime indicators. The authors use the clustering of subjects to show that, among the Federal Districts, the specific weight of the third class is prevalent in the Siberian Federal District in 2010–2020. The presented methodology can be applied to create a probability model of predicting that a subject or a District will be in a certain class.

Automatic Quasi-Clique Merger Algorithm — A hierarchical clustering based on subgraph-density

The Automatic Quasi-Clique Merger algorithm is a new algorithm adapted from early work published under the name QCM (introduced by Ou and Zhang (2007)). The AQCM algorithm performs hierarchical clustering in any data set for which there is an associated similarity measure quantifying the similarity of any data i and data j. Importantly, the method exhibits two valuable performance properties: (1) the ability to automatically return either a larger or smaller number of clusters depending on the inherent properties of the data rather than on a parameter. (2) the ability to return a very large number of relatively small clusters automatically when such clusters are reasonably well defined in a data set. In this work we present the general idea of a quasi-clique agglomerative approach, provide the full details of the mathematical steps of the AQCM algorithm, and explain some of the motivation behind the new methodology. The main achievement of the new methodology is that the agglomerative process now unfolds adaptively according to the inherent structure unique to a given data set, and this happens without the time-costly parameter adjustment that drove the previous QCM algorithm. For this reason we call the new algorithm automatic. We provide a demonstration of the algorithm’s performance at the task of community detection in a social media network of 22,900 nodes.

Features and diversity of symptoms of moyamoya disease in adolescents: A cluster analysis.

The purpose of this study is to explore the symptoms experienced by adolescents with moyamoya disease and to identify the characteristics of each symptom cluster associated with moyamoya disease. A retrospective and descriptive design, which is a secondary data analysis study based on electronic medical record data from hospitals. To assess the symptoms associated with moyamoya disease, a qualitative study was conducted on 12 adolescents, 12 caregivers and 12 experts on moyamoya disease. According to a qualitative study, 12 main symptoms (limb weakness, numbness, pins and needles, difficulty in speech, blurred vision, facial paralysis, hand tremors, involuntary movements, convulsions, dizziness, nausea/vomiting and headaches) were selected. Data were collected around these symptoms. The data collection was carried out through the Electronic Medical Record (EMR) data of 303 adolescents aged 13-19 who were diagnosed with moyamoya disease (ICD 10 Code: I67.5) between January 2010-December 2019 in a hospital in Seoul, South Korea. Cluster analysis was applied to identify symptom clusters with the hierarchical cluster agglomerative approach. We used the log-likelihood distance to measure the similarity of variables. Proximity between groups of variables was measured using the two-step method. The physical symptoms experienced by adolescents with moyamoya disease were 'limb weakness' - the most common - followed by a 'headache,' 'difficulty in speech,' and 'nausea/vomiting.' A total of five symptom clusters were derived: cluster 1 was characterized by 'limb weakness'; cluster 2 participants were asymptomatic or experienced 'convulsions'; cluster 3 experienced 'difficulty in speech' and 'facial paralysis'; cluster 4 is prone to 'dizziness' and 'pins and needles'; and cluster 5 displays 'headaches' and 'nausea/vomiting'. This study offers a multidimensional approach for identifying differences in clinical symptoms of moyamoya disease among adolescents. These results will help provide interventions concerning the characteristics of the symptoms of moyamoya disease among adolescents.

Agglomerative Info-Clustering: Maximizing Normalized Total Correlation

We show that, under the info-clustering framework, correlated random variables can be clustered in an agglomerative manner. While the existing divisive approach successively segregates the random variables into subsets with increasing multivariate mutual information, our agglomerative approach successively merges subsets of random variables sharing a large amount of normalized total correlation. We show that both approaches result in the same hierarchy of clusters, but the agglomerative approach is an order of magnitude faster than the divisive one. The uniqueness of the hierarchy produced by the two approaches is due to a fundamental connection that we uncover between the well-known total correlation and the recently proposed measure of multivariate mutual information. We implement the new algorithm and provide a data structure for efficient storage and retrieval of the hierarchical clustering solution.

A time series analysis-based stock price prediction using machine learning and deep learning models

Prediction of future movement of stock prices has always been a challenging task for the researchers. While the advocates of the efficient market hypothesis (EMH) believe that it is impossible to design any predictive framework that can accurately predict the movement of stock prices, there are seminal work in the literature that have clearly demonstrated that the seemingly random movement patterns in the time series of a stock price can be predicted with a high level of accuracy. Design of such predictive models requires choice of appropriate variables, right transformation methods of the variables, and tuning of the parameters of the models. In this work, we present a very robust and accurate framework of stock price prediction that consists of an agglomeration of statistical, machine learning and deep learning models. We use the daily stock price data, collected at five minutes interval of time, of a very well known company that is listed in the National Stock Exchange (NSE) of India. The granular data is aggregated into three slots in a day, and the aggregated data is used for building and training the forecasting models. We contend that the agglomerative approach of model building that uses a combination of statistical, machine learning, and deep learning approaches, can very effectively learn from the volatile and random movement patterns in a stock price data. We build eight classification and eight regression models based on statistical and machine learning approaches. In addition to these models, a deep learning regression model using a long-and-short-term memory (LSTM) network is also built. Extensive results have been presented on the performance of these models, and the results are critically analyzed.

Community detection in complex networks using a new agglomerative approach

Complex networks are used for the representation of complex systems such as social networks. Graph analysis comprises various tools such as community detection algorithms to uncover hidden data. Community detection aims to detect similar subgroups of networks that have tight interconnections with each other while, there is a sparse connection among different subgroups. In this paper, a greedy and agglomerative approach is proposed to detect communities. The proposed method is fast and often detects high-quality communities. The suggested method has several steps. In the first step, each node is assigned to a separated community. In the second step, a vertex is selected randomly and then its neighbors are determined. Then the selected node and its best neighbor will be merged if their merging brings positive gain. The merging of the selected vertex and its best neighbor has more gain than other neighbors. Whenever the merging occurs, the graph will be updated and this process will be continued until all the vertexes are assessed. Furthermore, the computational complexity of the proposed method is $O(nm)$, where $n$ and $m$ refer to the total number of vertexes and edges, respectively. In addition, our proposed method is compared with the Girvan--Newman algorithm and the fast divisive method for community detection. Results show that the proposed method is much faster than them and can detect high-quality communities. Finally, the accuracy of the proposed method is evaluated by using four different measures of purity, F-measure, NMI, and ARI.

Distinct asthma phenotypes with low maximal attainment of lung function on cluster analysis

Objective Asthma is a heterogeneous disease with varying clinical presentations, severity and ability to achieve asthma control. The present study aimed to characterize clinical phenotypes of asthma in an Indian cohort of subjects using a cluster analysis approach. Methods Patients with confirmed asthma (N = 100) and at least 6-months of follow-up data, identified by retrospective chart review, were included in this study. Demographics, age at disease onset, disease duration, body mass index, serial spirometry and allergen sensitization were assessed. Asthma control was assessed prospectively using Global Initiative for Asthma and Asthma Control Test. R version 3.4.3 was used for statistical analysis. Ward’s minimum-variance hierarchical clustering method was performed using an agglomerative (bottom-up) approach. To compare differences between clusters, analysis of variance using Kruskal-Wallis test (continuous variables) and chi-square test (categorical variables) was used. Results Cluster analysis of 100 treatment-naive patients with asthma identified four clusters. Cluster 1, (N = 40), childhood onset of disease, normal body weight, equal gender distribution and achieved normal lung function. Cluster 2 (N = 16) included adolescent disease-onset, obese, majority males and had poor attainment of maximum lung functions. Cluster 3 (N = 20) were older, late-onset of disease, obese, majority male and had poor attainment of maximum lung function. Cluster 4 (N = 24) had adult-onset of disease, obese, predominantly female and achieved normal lung function. Conclusions In an Indian cohort of well-characterized patients with asthma, cluster analysis identified four distinct clinical phenotypes of asthma, two of which had poor attainment of maximum lung functions.

A similarity based generalized modularity measure towards effective community discovery in complex networks

Modularity is a widely used goodness metric that effectively measures the strength of the community structures present in a network. However its performance may not be desirable for identifying densely connected communities or clusters of a network. It also often fails to identify communities or clusters that contain very few nodes. Furthermore, modularity is defined based only on the exact node-to-node connectivity of a network while disregarding their neighborhood connectivity. In this paper, we associate the neighborhood connectivity to the modularity function and propose a generalized modularity function based on the node similarity measure which quantifies the quality of a given network partition. Making use of this similarity based modularity function, an effective agglomerative approach for identifying communities is introduced. This agglomerative approach iteratively discovers the final community structure of the network by finding and merging together, at each step, the community pairs which maximize the proposed modularity value. A significant characteristic of the proposed method is that it does not need any prior knowledge about the actual communities of a network. The performance of the proposed method and state-of-the-art algorithms are compared using the value of modularity, normalized mutual information and adjusted variation of information measures on several real world and artificial networks. The empirical results show the effectiveness of the proposed method compared to the state-of-the-art techniques.

Topological and functional comparison of community detection algorithms in biological networks

BackgroundCommunity detection algorithms are fundamental tools to uncover important features in networks. There are several studies focused on social networks but only a few deal with biological networks. Directly or indirectly, most of the methods maximize modularity, a measure of the density of links within communities as compared to links between communities.ResultsHere we analyze six different community detection algorithms, namely, Combo, Conclude, Fast Greedy, Leading Eigen, Louvain and Spinglass, on two important biological networks to find their communities and evaluate the results in terms of topological and functional features through Kyoto Encyclopedia of Genes and Genomes pathway and Gene Ontology term enrichment analysis. At a high level, the main assessment criteria are 1) appropriate community size (neither too small nor too large), 2) representation within the community of only one or two broad biological functions, 3) most genes from the network belonging to a pathway should also belong to only one or two communities, and 4) performance speed. The first network in this study is a network of Protein-Protein Interactions (PPI) in Saccharomyces cerevisiae (Yeast) with 6532 nodes and 229,696 edges and the second is a network of PPI in Homo sapiens (Human) with 20,644 nodes and 241,008 edges. All six methods perform well, i.e., find reasonably sized and biologically interpretable communities, for the Yeast PPI network but the Conclude method does not find reasonably sized communities for the Human PPI network. Louvain method maximizes modularity by using an agglomerative approach, and is the fastest method for community detection. For the Yeast PPI network, the results of Spinglass method are most similar to the results of Louvain method with regard to the size of communities and core pathways they identify, whereas for the Human PPI network, Combo and Spinglass methods yield the most similar results, with Louvain being the next closest.ConclusionsFor Yeast and Human PPI networks, Louvain method is likely the best method to find communities in terms of detecting known core pathways in a reasonable time.

A contextual-based segmentation of compact PolSAR images using Markov Random Field (MRF) model

ABSTRACTAs the first major step in each object-oriented feature extraction approach, segmentation plays an essential role as a preliminary step towards further and higher levels of image processing. The primary objective of this paper is to illustrate the potential of Polarimetric Synthetic Aperture Radar (PolSAR) features extracted from Compact Polarimetry (CP) SAR data for image segmentation using Markov Random Field (MRF). The proposed method takes advantage of both spectral and spatial information to segment the CP SAR data. In the first step of the proposed method, k-means clustering was applied to over-segment the image using the appropriate features optimally selected using Genetic Algorithm (GA). As a similarity criterion in each cluster, a probabilistic distance was used for an agglomerative hierarchical merging of small clusters into an appropriate number of larger clusters. In the agglomerative clustering approach, the estimation of the appropriate number of clusters using the data log-likelihood algorithm differs depending on the distance criterion used in the algorithm. In particular, the Wishart Chernoff distance which is independent of samples (pixels) tends to provide a higher appropriate number of clusters compared to the Wishart test statistic distance. This is because the Wishart Chernoff distance preserves detailed data information corresponding to small clusters. The probabilistic distance used in this study is Wishart Chernoff distance which evaluates the similarity of clusters by measuring the distance between their complex Wishart probability density functions. The output of this step, as the initial segmentation of the image, is applied to a Markov Random Field model to improve the final segmentation using vicinity information. The method combines Wishart clustering and enhanced initial clusters in order to access the posterior MRF energy function. The contextual image classifier adopts the Iterated Conditional Mode (ICM) approach to converge to a local minimum and represent a good trade-off between segmentation accuracy and computation burden. The results showed that the PolSAR features extracted from CP mode can provide an acceptable overall accuracy in segmentation when compared to the full polarimetry (FP) and Dual Polarimetry (DP) data. Moreover, the results indicated that the proposed algorithm is superior to the existing image segmentation techniques in terms of segmentation accuracy.

FuzAg: Fuzzy Agglomerative Community Detection by Exploring the Notion of Self-Membership

In this paper, a fuzzy agglomerative (FuzAg) approach is proposed for community detection that iteratively updates membership degree of nodes. Earlier approaches assign membership degree to nodes based on communities only. We introduce the notion of self-membership in addition to the membership of different communities. The essence of self-membership is to give opportunity to all nodes in growing their own community. Nodes having higher self-membership degree are referred as anchors , and they get a chance to expand their associated community. Meanwhile, some new anchors may emerge in successive iterations, whereas false or redundant anchors get removed. The time complexity of the proposed algorithm is shown to be $O(n^2)$ . We compare the results of the proposed FuzAg algorithm with those of state-of-the-art fuzzy community detection algorithms on ten real-world datasets as well as on synthetic networks. Results indicated by various quality and accuracy metrics show impressive performance of FuzAg in identifying both disjoint communities and fuzzy communities.

Supervoxel-based segmentation of 3D imagery with optical flow integration for spatiotemporal processing

The past 20 years has seen a progressive evolution of computer vision algorithms for unsupervised 2D image segmentation. While earlier efforts relied on Markov random fields and efficient optimization (graph cuts, etc.), the next wave of methods beginning in the early part of this century were, in the main, stovepiped. Of these 2D segmentation efforts, one of the most popular and, indeed, one that comes close to being a state of the art method is the ultrametric contour map (UCM). The pipelined methodology consists of (i) computing local, oriented responses, (ii) graph creation, (iii) eigenvector computation (globalization), (iv) integration of local and global information, (v) contour extraction, and (vi) superpixel hierarchy construction. UCM performs well on a range of 2D tasks. Consequently, it is somewhat surprising that no 3D version of UCM exists at the present time. To address that lack, we present a novel 3D supervoxel segmentation method, dubbed 3D UCM, which closely follows its 2D counterpart while adding 3D relevant features. The methodology, driven by supervoxel extraction, combines local and global gradient-based features together to first produce a low-level supervoxel graph. Subsequently, an agglomerative approach is used to group supervoxel structures into a segmentation hierarchy with explicitly imposed containment of lower-level supervoxels in higher-level supervoxels. Comparisons are conducted against state of the art 3D segmentation algorithms. The considered applications are 3D spatial and 2D spatiotemporal segmentation scenarios. For the latter comparisons, we present results of 3D UCM with and without optical flow video pre-processing. As expected, when motion correction beyond a certain range is required, we demonstrate that 3D UCM in conjunction with optical flow is a very useful addition to the pantheon of video segmentation methods.