Multi-scale Clustering Research Articles

Forecasting energy spot prices: A multiscale clustering recognition approach

Effective and timely forecasting of energy spot price plays a key role in the international bulk commodities market, which is related to the goal and direction of the reform of pricing mechanism. Previous studies are based on the decomposition-ensemble model to forecast the time series of high volatility in energy spot price, but ignore the adaptability of different models for different sub-series. This paper presents a hybrid forecasting model based on multiscale clustering recognition approach. Data preprocessing module is used to split the original series into several sub-series to reduce the complexity of the data, and then the optimal mode of input data is determined by feature selection. To identify the characteristic of sub-series, fuzzy cluster approach is introduced to distinguish with their complexity, which can divide adaptively them into different components with different frequency. This divide and conquer strategy give the reference for selecting suitable forecasting engine. With the help of tuned forecasting engines, these components are conducted respective prediction and the final results are reconfigured with the existing time point. The proposed model takes fully the advantages of different models into account, which ensures to use the data effectively. Through the empirical results, it demonstrates that the multiscale clustering recognition approach can both produce satisfactory prediction which is close to the actual values and be regarded as a promising tool for forecasting energy spot price.

Multi-Scale Massive Points Fast Clustering Based on Hierarchical Density Spanning Tree

Spatial clustering is dependent on spatial scales. With the widespread use of web maps, a fast clustering method for multi-scale spatial elements has become a new requirement. Therefore, to cluster and display elements rapidly at different spatial scales, we propose a method called Multi-Scale Massive Points Fast Clustering based on Hierarchical Density Spanning Tree. This study refers to the basic principle of Clustering by Fast Search and Find of Density Peaks aggregation algorithm and introduces the concept of a hierarchical density-based spanning tree, combining the spatial scale with the tree links of elements to propose the corresponding pruning strategy, and finally realizes the fast multi-scale clustering of elements. The first experiment proved the time efficiency of the method in obtaining clustering results by the distance-scale adjustment of parameters. Accurate clustering results were also achieved. The second experiment demonstrated the feasibility of the method at the aggregation point element and showed its visual effect. This provides a further explanation for the application of tree-link structures.

Research on deep mining model of online learning data based on multiscale clustering

Research on deep mining model of online learning data based on multiscale clustering

Deep mining of mobile learning data based on multi-scale clustering analysis

Deep mining of mobile learning data based on multi-scale clustering analysis

Wall-bounded thermal turbulent convection driven by heat-releasing point particles

In this work we investigate the thermal convection driven by heat-releasing point particles. Three-dimensional direct numerical simulations are conducted for$1\times 10^7\le {\textit {Ra}} \le 1\times 10^{10}$and$0.01\le {\textit {St}} \le 10$, where the Rayleigh number${\textit {Ra}}$and Stokes number${\textit {St}}$measure the strengths of the heat releasing rate and the Stokes drag, respectively. A regime at intermediate Stokes numbers is identified with most particles accumulating into the top boundary layer region, while for other cases particles are constantly advected over the entire domain. For the latter state, the flow motions are stronger at the upper part of the domain. The thicknesses of both momentum and thermal boundary layers at the top plate follow the same scaling law with${\textit {Ra}}$and show minor dependences on${\textit {St}}$. The volume-averaged temperature and convective flux exhibit non-monotonic variations as${\textit {St}}$increases and reach their minimums at intermediate${\textit {St}}$. The fraction coefficient of heat flux, i.e. the ratio between the heat flux through the bottom plate and the total flux through both plates, shares the similar dependence on${\textit {St}}$as the convective flux. The relation between these scaling laws can be explained by using the global balance between the dissipation and convective flux. The scaling laws for the transition between different flow regimes are also proposed and agree with the numerical results. The preferential concentration of particles is observed for all cases and is strongest at intermediate Stokes numbers, for which multiscale clustering emerges with small clusters forming larger ones.

Spatiotemporal dynamic evolution and influencing factors of family farms in urban agglomerations in the middle reaches of the Yangtze River

Family farms, considered the most desirable form of Chinese agriculture, play a pivotal role in promoting rural revitalization and agricultural modernization. The purpose of this study was to summarize the spatiotemporal evolution characteristics and influencing factors of family farms to better promote the development of modern agriculture. Using provincial demonstration family farms in the urban agglomeration in the middle reaches of the Yangtze River (MYR-UA) as the research object, this study applied the nearest neighbor index, kernel density analysis, multiscale spatial clustering analysis (Ripley’s K-function), and geographically weighted regression (GWR) model to reveal the spatiotemporal dynamic evolution and influencing factors of family farms. The results indicate that: 1) from 2013 to 2021, family farms exhibited annual increases, and their development stages could be divided into rapid, stable, and slow growth periods. 2) The spatial agglomeration pattern of family farms was significant, and the intercepted points at different time periods show the distribution characteristics of the entire dispersion and local concentration. The spatial evolution characteristics of different types of family farms are nearly consistent with those of the overall family farms. 3) The overall family farms and various types of family farms show a scale effect, which first strengthens and then weakens with the change in geographical distance. 4) The spatial pattern of family farms in MYR-UA is affected by both natural and social factors, of which, social factors had the greatest influence. Finally, based on the findings of the study, policy recommendations for promoting the high-quality development of family farms are proposed.

Quantitative assessment and comparison of urban patterns in Germany and the United States

In this paper, we present methods to assess the homogeneity of the settlement landscape and to identify settlement clusters in terms of location and size. Due to various input data, methodologies, and spatial concepts, there are only a few international comparative research studies with quantitative, spatial approaches to date. Existing studies are mostly qualitative and empirically hardly replicable. Here, we introduce two novel methodological approaches to describe urban land patterns in a numerically stable, consistent and globally comparable manner. The first method evaluates the multi-scale homogeneity of the settlements and the respective density distributions in a purely non-parametric approach. The second method transfers the settlement patterns into a Gaussian Mixture Model via hierarchical multi-scale clustering in order to describe each cluster by the parameters of a 2D Gaussian distribution. In order to proof the robustness of both approaches, two different reference units are considered for the interpretation of the results: OECD functional urban areas and standardized subsets of 200 km × 200 km around the same central cities. We show that German urban areas are characterized by a higher heterogeneity within a smaller neighbourhood with an almost symmetric density distribution in comparison to urban areas in the U.S. Furthermore, German urban areas possess an increasing homogeneity with increasing size, i.e., the larger the urban agglomeration, the more uniform it appears. It is the contrary in the U.S. These findings are substantiated by the analysis of individual city centres. In Germany, many clusters of similar size define the urban region, whereas in the U.S. one large dominating cluster exceeds most neighbouring settlements significantly. The study is performed on the Global Urban Footprint Density, which assigns a density value to each urban pixel of about 30 m by 30 m. In this way, the approach is a blueprint to be extended to urban patterns of any spatial unit worldwide.

Multi-scale deep multi-view subspace clustering with self-weighting fusion and structure preserving

Deep subspace clustering methods for multi-view have achieved impressive clustering performance over other clustering methods. However, the existing methods either cannot integrate the global and local information of multi-view or fail to explore the discriminative contributions among views. In this paper, we propose a novel multi-scale deep multi-view subspace clustering (MDMVSC) method, which unifies the multi-scale learning (ML) module, self-weighting fusion (SF) module and structure preserving (SP) constraint. Specifically, to take advantage of the complementarity and diversity of different views, ML module first learns specific self-representation matrix for each view from the multi-scale low-dimensional latent features with the global and local information. Then, using the SF module, these matrices are fused to obtain the consensus representation of multi-view via attention mechanism guided weights according to their discriminative contributions. Moreover, SP constraint encourages the multi-scale latent features to preserve the consistent structural information of the original multi-view for enhancing representation ability. Extensive experimental results on five datasets demonstrate the superiority of MDMVSC in comparison with several state-of-the-art methods.

Droplet cluster evolution and collective gasification of droplet groups in a fuel spray: A comparative study under non-reacting and reacting conditions

In this paper, the goal is to understand the influence of spray combustion on the evolution of droplet clusters and collective gasification processes of droplet groups in a fuel spray. Experiments were conducted in an acetone spray from a pressure swirl injector under both non-reacting and reacting conditions for the same fuel flow rate through the injector. The PIV technique was employed to capture Mie-scattering images of the spray and to measure droplet velocity, while droplet sizing was achieved by application of the ILIDS technique. Voronoi analysis of the PIV images facilitated identification of droplet clusters. Accordingly, the cluster length scale and local droplet number density within clusters could be obtained. In addition, the Group evaporation/combustion number (G) for individual clusters was also evaluated. Either in the presence or absence of spray combustion, wide range of cluster size exists in the spray which highlights multi-scale clustering of droplets. Accordingly, G varies in a broad range that indicates multi-mode evaporation/combustion of the droplet clusters. Distinct behavior of droplet clusters is identified depending on the cluster size relative to the mean cluster area. While the evolution of small clusters is not sensitive to the presence of the spray flame, the length scale of the large-scale clusters and gasification process of such groups of droplets are modified due to spray combustion compared to the non-reacting condition.

A novel solution of deep learning for endoscopic ultrasound image segmentation: enhanced computer aided diagnosis of gastrointestinal stromal tumor

Gastrointestinal stromal tumor is one of the critical tumors that doctors do not suggest to get frequent endoscopy, so there is a need for a diagnosis system which can process ultrasound images and figure out the tumor. Many gastrointestinal tumor diagnosis methods were developed, but all of these methods used manual contour rather than automatic segmentation. The research adopts enhanced automatic segmentation to improve the diagnosis of the gastrointestinal stromal tumor with deep convolutional neural networks. This solution’s proposed system is an enhanced automated segmentation methodology using multi-scale Gaussian kernel fuzzy clustering and multi-scale vector field convolution, which segments the ultrasound image automatically into the region of interest (the infected area). Convolutional Neural Network with Class Activation Mapping is done to diagnose an image with the tumor for Four datasets, namely (USS1, SH Hospital, SNUH, BUSI). This proposed system helps to get a clearer tumor image, and the accuracy has increased from 84.275% to 88.4%, and the processing time has reduced from 28.525% to 24.575%. The proposed solution enhanced Automatic Segmentation helped to get clearer tumor image which resulted in increased accuracy and decreased performance time compared to the state-of-the-art. Automatic segmentation overcomes the dependency on the expert for drawing the Region of Interest (ROI).

A multiscale environment for learning by diffusion

• We introduce the MELD data model: a diffusion framework for multiscale clustering. • We show how cluster coherence and separation interact with diffusion in MELD clusterings. • We introduce the M-LUND multiscale clustering algorithm and guarantee its performance. • We guarantee that M-LUND recovers the MELD data model from many datasets. • We show M-LUND extracts latent multiscale structure in synthetic and real datasets. Clustering algorithms partition a dataset into groups of similar points. The clustering problem is very general, and different partitions of the same dataset could be considered correct and useful. To fully understand such data, it must be considered at a variety of scales, ranging from coarse to fine. We introduce the Multiscale Environment for Learning by Diffusion (MELD) data model, which is a family of clusterings parameterized by nonlinear diffusion on the dataset. We show that the MELD data model precisely captures latent multiscale structure in data and facilitates its analysis. To efficiently learn the multiscale structure observed in many real datasets, we introduce the Multiscale Learning by Unsupervised Nonlinear Diffusion (M-LUND) clustering algorithm, which is derived from a diffusion process at a range of temporal scales. We provide theoretical guarantees for the algorithm's performance and establish its computational efficiency. Finally, we show that the M-LUND clustering algorithm detects the latent structure in a range of synthetic and real datasets.

Multi-scale semi-supervised clustering of brain images: Deriving disease subtypes.

Disease heterogeneity is a significant obstacle to understanding pathological processes and delivering precision diagnostics and treatment. Clustering methods have gained popularity for stratifying patients into subpopulations (i.e., subtypes) of brain diseases using imaging data. However, unsupervised clustering approaches are often confounded by anatomical and functional variations not related to a disease or pathology of interest. Semi-supervised clustering techniques have been proposed to overcome this and, therefore, capture disease-specific patterns more effectively. An additional limitation of both unsupervised and semi-supervised conventional machine learning methods is that they typically model, learn and infer from data using a basis of feature sets pre-defined at a fixed anatomical or functional scale (e.g., atlas-based regions of interest). Herein we propose a novel method, "Multi-scAle heteroGeneity analysIs and Clustering" (MAGIC), to depict the multi-scale presentation of disease heterogeneity, which builds on a previously proposed semi-supervised clustering method, HYDRA. It derives multi-scale and clinically interpretable feature representations and exploits a double-cyclic optimization procedure to effectively drive identification of inter-scale-consistent disease subtypes. More importantly, to understand the conditions under which the clustering model can estimate true heterogeneity related to diseases, we conducted extensive and systematic semi-simulated experiments to evaluate the proposed method on a sizeable healthy control sample from the UK Biobank (N=4403). We then applied MAGIC to imaging data from Alzheimer's disease (ADNI, N=1728) and schizophrenia (PHENOM, N=1166) patients to demonstrate its potential and challenges in dissecting the neuroanatomical heterogeneity of common brain diseases. Taken together, we aim to provide guidance regarding when such analyses can succeed or should be taken with caution. The code of the proposed method is publicly available at https://github.com/anbai106/MAGIC.

Research on deep mining model of online learning data based on multiscale clustering

Research on deep mining model of online learning data based on multiscale clustering

A Multiscale Spectral Features Graph Fusion Method for Hyperspectral Band Selection

This article proposes a multiscale spectral features graph fusion (MSFGF) method for selecting proper hyperspectral bands. The MSFGF regards that the selected bands should reflect diagnostic spectral information of ground objects at different scales, and it explores band selection from the aspect of multiple spatial scales. First, it adopts the multiscale low-rank decomposition (MSLRD) model to find multiscale spectral features of different ground objects. The model considers divergent spatial structures or spatial correlations of ground objects at different scales, and factorizes the hyperspectral data cube into a series of low-rank block-wise data cubes, where the blocks take spatial structures of different ground objects at increasing scales. Second, the MSFGF presents the multiscale sparse spectral clustering (MSSC) model to fuse the separate connected graphs of multiscale spectral features into a consensus graph. The consensus graph combines the complementary information of multiscale spectral features and helps to reveal the intrinsic clustering structure of all spectral bands. Finally, the MSFGF utilizes spectral clustering to find clusters from the consensus graph and selects representative bands. Experimental results on three widely used hyperspectral data prove the superiority of MSFGF in selecting bands, where it outperforms other seven state-of-the-art methods in classification with an acceptable computational cost.

Superpixel-Guided Variable Gabor Phase Coding Fusion for Hyperspectral Image Classification

3-D Gabor, as a typical filter, plays a critical role in extracting discriminative spectral–spatial features from hyperspectral images (HSIs). However, the performance of traditional 3-D Gabor is limited by the uniform response to each direction, which is inconsistent with the complexity of land cover distribution. It has been a continuing concern for researchers to investigate the anisotropic 3-D Gabor filters. In addition, the 3-D Gabor wavelets do not make full use of spatial distribution information, thus reducing the accuracy. This article proposes a superpixel-guided variable 3-D Gabor phase coding fusion ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Su</i> VGF) framework for HSI classification with limited training samples. First, the variable 3-D Gabor filters are created based on various asymmetric sinusoidal waves and spatial kernel sizes to achieve multidirectional features. Second, the local Gabor phase ternary pattern is adopted to encode the Gabor phases and improve the feature discrimination. Meanwhile, a scale map is produced by the majority voting of multiscale simple noniterative clustering (SNIC) and entropy rate superpixel (ERS) segmentation, which contains sufficient and complementary spatial distribution information. Then, geometric optimization is employed on the scale map to reduce noise disturbances. Finally, all Gabor features are modified by the filter with the guidance of a scale map and fused together as a confidence cube, and the random forest algorithm is exploited for classification. The <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Su</i> VGF is applied to three real hyperspectral datasets to demonstrate the superiority of higher accuracy, stronger robustness, and less computational complexity in comparison with several state-of-the-art ones.

Multi-Scale Deep Subspace Clustering With Discriminative Learning

Deep subspace clustering methods have achieved impressive clustering performance compared with other clustering algorithms. However, most existing methods suffer from the following problems: 1) they only consider the global features but neglect the local features in subspace self-expressiveness learning; 2) they neglect the discriminative information of each self-expressiveness coefficient matrix; 3) they ignore the useful long-range dependencies and positional information in feature representation learning. To solve these problems, in this paper, we propose a novel multi-scale deep subspace clustering with discriminative learning (MDSCDL) to obtain a high-quality self-expressiveness coefficient matrix. Specifically, MDSCDL bridges multiple fully-connection layers between encoder and decoder to learn multi-scale self-expressiveness coefficient matrices from global and local features, representing the more comprehensive relationship among data. By modeling the interdependencies of the multi-scale self-expressiveness coefficient matrices, MDSCDL adaptively assigns discriminative weights for each matrix and fuses them with convolution operation. Moreover, to increase representation power, MDSCDL introduces the coordinate attention mechanism to extract the long-range dependencies and positional features for subspace self-expressiveness learning. Extensive experiments on the face and object datasets have shown the superiority of MDSCDL compared with several state-of-the-art methods.

ClusterMap for multi-scale clustering analysis of spatial gene expression

Quantifying RNAs in their spatial context is crucial to understanding gene expression and regulation in complex tissues. In situ transcriptomic methods generate spatially resolved RNA profiles in intact tissues. However, there is a lack of a unified computational framework for integrative analysis of in situ transcriptomic data. Here, we introduce an unsupervised and annotation-free framework, termed ClusterMap, which incorporates the physical location and gene identity of RNAs, formulates the task as a point pattern analysis problem, and identifies biologically meaningful structures by density peak clustering (DPC). Specifically, ClusterMap precisely clusters RNAs into subcellular structures, cell bodies, and tissue regions in both two- and three-dimensional space, and performs consistently on diverse tissue types, including mouse brain, placenta, gut, and human cardiac organoids. We demonstrate ClusterMap to be broadly applicable to various in situ transcriptomic measurements to uncover gene expression patterns, cell niche, and tissue organization principles from images with high-dimensional transcriptomic profiles.

A bearing fault diagnosis method based on multiscale dispersion entropy and GG clustering

Fault diagnosis of rolling bearings depends on the construction of an effective index and a reasonable identification method of fault features. In this paper, an effective method to identify fault type and degree is presented. Firstly, each element in the training samples is decomposed by ensemble empirical mode decomposition, and then the dispersion entropy of the intrinsic mode function is calculated to construct the feature matrix of the training samples. Secondly, principal component analysis visually reduces the dimension of the feature matrix, and the Gath-Geva clustering method divides the reduced two-dimensional matrix to obtain the clustering center and category of the training sample features. Finally, the normalized clustering distance between the feature matrix of test samples and the clustering center of training samples is used to judge the membership. The effectiveness of the proposed method was verified by the Case Western Reserve University (CWRU) data set, the QPZZ-II platform, and Cincinnati Intelligent Maintenance Systems.

Multi-scale graph attention subspace clustering network

Deep subspace clustering (DSC) network which uses deep auto-encoders (DAE) mapping raw data into a latent space for clustering has achieved significant performance. However, when it comes to graph-based clustering, it fails to encode the node attribute and graph structure simultaneously, which degrades the clustering performance. Meanwhile, existing graph convolutional network (GCN) based clustering methods usually do not have a clustering-oriented objection function since the process of learning representation and clustering are separated. In addition, both aforementioned methods can neither catch the multi-scale information nor use the current clustering labels effectively, which lead to the suboptimal performance. To this end, we proposed a novel end-to-end framework called Multi-Scale Graph Attention Subspace Clustering Network (MSGA). By employing a novel GCN-based feature extraction module, it can effectively capture the node representation on graph-based datasets. Moreover, a multi-scale self-expression module is designed for obtaining a more discriminative coefficient representation from each layer of the encoder and a self-supervised module is introduced for supervising the learning of node representation. Specifically, we train and optimize them in a unified framework so that different modules can benefit from each other. Extensive experiments demonstrate the superiority of our method compared with several state-of-the-art methods.

MNHN-Tree-Tools: a toolbox for tree inference using multi-scale clustering of a set of sequences.

Genomic sequences are widely used to infer the evolutionary history of a given group of individuals. Many methods have been developed for sequence clustering and tree building. In the early days of genome sequencing, these were often limited to hundreds of sequences but due to the surge of high throughput sequencing, it is now common to have millions of sampled sequences at hand. We introduce MNHN-Tree-Tools, a high performance set of algorithms that builds multi-scale, nested clusters of sequences found in a FASTA file. MNHN-Tree-Tools does not rely on multiple sequence alignment and can thus be used on large datasets to infer a sequence tree. Herein, we outline two applications: a human alpha-satellite repeats classification and a tree of life derivation from 16S/18S rDNA sequences. Open source with a Zlib License via the Git protocol: https://gitlab.in2p3.fr/mnhn-tools/mnhn-tree-tools. A detailed users guide and tutorial: https://gitlab.in2p3.fr/mnhn-tools/mnhn-tree-tools-manual/-/raw/master/manual.pdf. http://treetools.haschka.net. Supplementary data are available at Bioinformatics online.