Classical Clustering Algorithms Research Articles

Clustering time series by extremal dependence

AbstractThe goal of this paper is to characterize the temporal dependence structure on the extremes of time series and use such dependency to group them. In particular, three similarity measures to capture extremal dependence are proposed, being their performance assessed in different scenarios. This will involve the use of classical time series clustering algorithms, as well as rigorous evaluation of their performance in both simulated scenarios and real-world time series data sets. The focus will be on comparing the performance of these similarity measures with different clustering methods, and illustrate the efficacy of extremal dependence-based clustering in meteorological data. To achieve this, we will consider a dataset consisting of daily maximum temperatures recorded at 500 stations across Europe.

Classification of synoptic circulation patterns with a two-stage clustering algorithm using the modified structural similarity index metric (SSIM)

Abstract. We develop a new classification method for synoptic circulation patterns with the aim to extend the evaluation routine for climate simulations. This classification is applicable to any region of the globe of any size given the reference data. Its unique novelty is the use of the modified structural similarity index metric (SSIM) instead of traditional distance metrics for cluster building. This classification method combines two classical clustering algorithms used iteratively, hierarchical agglomerative clustering (HAC) and k-medoids, with only one pre-set parameter – the threshold on the similarity between two synoptic patterns expressed as the structural similarity index measure (SSIM). This threshold is set by the user to imitate the human perception of the similarity between two images (similar structure, luminance, and contrast), whereby the number of final classes is defined automatically. We apply the SSIM-based classification method to the geopotential height at the pressure level of 500 hPa from the ERA-Interim reanalysis data for 1979–2018 and demonstrate that the built classes are (1) consistent with the changes in the input parameter, (2) well-separated, (3) spatially stable, (4) temporally stable, and (5) physically meaningful. We demonstrate an exemplary application of the synoptic circulation classes obtained with the new classification method for evaluating Coupled Model Intercomparison Project Phase 6 (CMIP6) historical climate simulations and an alternative reanalysis (for comparison purposes): output fields of CMIP6 simulations (and of the alternative reanalysis) are assigned to the classes and the Jensen–Shannon distance is computed for the match in frequency, transition, and duration probabilities of these classes. We propose using this distance metric to supplement a set of commonly used metrics for model evaluation.

A computational topology-based spatiotemporal analysis technique for honeybee aggregation

A primary challenge in understanding collective behavior is characterizing the spatiotemporal dynamics of the group. We employ topological data analysis to explore the structure of honeybee aggregations that form during trophallaxis, which is the direct exchange of food among nestmates. From the positions of individual bees, we build topological summaries called CROCKER matrices to track the morphology of the group as a function of scale and time. Each column of a CROCKER matrix records the number of topological features, such as the number of components or holes, that exist in the data for a range of analysis scales, at a given point in time. To detect important changes in the morphology of the group from this information, we first apply dimensionality reduction techniques to these matrices and then use classic clustering and change-point detection algorithms on the resulting scalar data. A test of this methodology on synthetic data from an agent-based model of honeybees and their trophallaxis behavior shows two distinct phases: a dispersed phase that occurs before food is introduced, followed by a food-exchange phase during which aggregations form. We then move to laboratory data, successfully detecting the same two phases across multiple experiments. Interestingly, our method reveals an additional phase change towards the end of the experiments, suggesting the possibility of another dispersed phase that follows the food-exchange phase.

ZSGEEHCP: Zonal Stable Gateway Based Energy Efficient Heterogeneous Clustering Protocol For WSN- Based IoT.

In future most of the devices on the earth connected to the internet for making world of intelligent networks. The internet of things (IoT) and Industrial IoT play a very significant role for implementing such kind of smart networks. So, in that case, the transmission of packets of data sent to the base station from sensor nodes with minimum energy consumption in a homogeneous or heterogeneous environment to hike the stability of the network appears as a big challenge for IoT-based networks. But, such kind of problem can be overcome by using a proper clustering algorithm. In this work, different kinds of classical clustering algorithms are studied precisely. Most of the traditional clustering algorithm protocols presume that all the nodes have the same amount of energy, they are unable to fully exploit the presence of node heterogeneity and the same problem will be faced with IoT-based networks. In this specific piece of work, we recommend a gateway based heterogeneous energy-based clustering algorithm which is crucial for many wireless sensor networks along with IoT-based applications. In this concept geographical area is divided into three zone, Advanced, Intermediate and normal zone. Here, advance nodes send data through gateway to the base station while intermediate node use clustering algorithm for transferring data and normal nodes communicate data to the base station directly. ZSGHCP is driven by the mounded election protocol probabilities of each node elected as a cluster head depending on the rest of the energy. By simulation, we show that ZSGHCP invariably extends the stability time as compared to current clustering procedures. Finally, we investigate the sensitivity of our ZSGHCP protocol to network heterogeneity characteristics that capture energy imbalance, throughput, and no of alive nodes. We construct that ZSGHCP payout extends the durability region for excessive values of additional energy brought by extra higher-powered nodes.

A scalable multi-density clustering approach to detect city hotspots in a smart city

In the field of Smart City applications, the analysis of urban data to detect city hotspots, i.e., regions where urban events (such as pollution peaks, virus infections, traffic spikes, and crimes) occur at a higher density than in the rest of the dataset, is becoming a common task. The detection of such hotspots can serve as a valuable organizational technique for framing detailed information about a metropolitan area, providing high-level spatial knowledge for planners, scientists, and policymakers. From the algorithmic viewpoint, classic density-based clustering algorithms are very effective in discovering hotspots characterized by homogeneous density; however, their application on multi-density data can produce inaccurate results. For such a reason, since metropolitan cities are characterized by areas with significantly variable densities, multi-density clustering approaches are more effective in discovering city hotspots. Moreover, the growing volumes of data collected in urban environments require the development of parallel approaches, in order to take advantage of scalable executions offered by Edge and Cloud environments. This paper describes the design and implementation of a parallel multi-density clustering algorithm aimed at analyzing high volumes of urban data in an efficient way. The experimental evaluation shows that the proposed parallel clustering approach takes out encouraging advantages in terms of execution time, speedup, and efficiency.

Pass-Efficient Algorithms for Graph Spectral Clustering (Student Abstract)

Graph spectral clustering is a fundamental technique in data analysis, which utilizes eigenpairs of the Laplacian matrix to partition graph vertices into clusters. However, classical spectral clustering algorithms require eigendecomposition of the Laplacian matrix, which has cubic time complexity. In this work, we describe pass-efficient spectral clustering algorithms that leverage recent advances in randomized eigendecomposition and the structure of the graph vertex-edge matrix. Furthermore, we derive formulas for their efficient implementation. The resulting algorithms have a linear time complexity with respect to the number of vertices and edges and pass over the graph constant times, making them suitable for processing large graphs stored on slow memory. Experiments validate the accuracy and efficiency of the algorithms.

Evaluasi Perbandingan Kinerja Algoritma Cheng and Church Biclustering Terhadap Algoritma Clustering Klasik K-Means untuk Mengidentifikasi Pola Distribusi Barang Ekspor Indonesia

Clustering is a process of grouping data into several groups (clusters) so that data in one cluster has a homogeneous level of similarity and data between clusters has heterogeneous similarity. A common example of a clustering algorithm is K-Means Clustering. Compared with classical clustering algorithms, the biclustering algorithm is a two-dimensional data grouping process. The biclustering algorithm functions to find data submatrices, namely row subgroups and column subgroups that have high correlation. One example of a biclustering algorithm is Cheng and Church Biclustering (CC Biclustering). The aim of this research is to evaluate the performance of the biclustering algorithm against classical clustering algorithms. Analysis applied to CC Biclustering and K-Means Clustering to identify distribution patterns of Indonesian export goods in the period 2013 to 2022. Based on research results, the optimal scenario for the K-Means algorithm is scenario 2, that is the application of the 7 cluster K-Means algorithm with pre- processing data scaling. Meanwhile, the optimal scenario for the CC Biclustering algorithm is scenario 1, that is the application of the CC Biclustering algorithm with a tolerance value of 0.10 with data scaling pre-processing. However, from these two scenarios, based on the MSR/Volume value, it was concluded that the best scenario is scenario 1 in the application of the CC Biclustering algorithm which has an MSR/Volume value of 0.077.

Functional-Coefficient Quantile Regression for Panel Data with Latent Group Structure

This article considers estimating functional-coefficient models in panel quantile regression with individual effects, allowing the cross-sectional and temporal dependence for large panel observations. A latent group structure is imposed on the heterogeneous quantile regression models so that the number of nonparametric functional coefficients to be estimated can be reduced considerably. With the preliminary local linear quantile estimates of the subject-specific functional coefficients, a classic agglomerative clustering algorithm is used to estimate the unknown group structure and an easy-to-implement ratio criterion is proposed to determine the group number. The estimated group number and structure are shown to be consistent. Furthermore, a post-grouping local linear smoothing method is introduced to estimate the group-specific functional coefficients, and the relevant asymptotic normal distribution theory is derived with a normalization rate comparable to that in the literature. The developed methodologies and theory are verified through a simulation study and showcased with an application to house price data from U.K. local authority districts, which reveals different homogeneity structures at different quantile levels.

Optimization Methods of Multi-Core Embedded System

With the development of the Internet of everything technology, embedded system has become one of the most common computing systems. Embedded system has high portability, but there are often stronger limitations in energy consumption, real-time and so on. In this work, have improved some traditional optimization algorithms, and finally get a task scheduling sequence, which can reduce the total execution time and cost of the task. Firstly, the multi-module division is used to divide multiple tasks into different modules, improve the classical Kernighan-Lin (KL) algorithm and clustering algorithm implement this process. Second, this paper invokes a series of algorithms to calculate the priority value of the task. Finally, this paper call two multi-core scheduling algorithms to schedule the tasks within each module to the Central Processing Unit (CPU) in the module. The proposed algorithm is implemented in this paper. With directed acyclic graph as input, the improvement effect of the algorithm and the direct advantages and disadvantages of different algorithms are evaluated with different task quantity.

The Path of Integrating Curriculum Civics into Advertising Design Teaching in Higher Vocational Colleges and Universities Based on Big Data Analysis

Abstract In order to cultivate advertising design applied talents, this paper firstly combs through the status quo of the current teaching system of advertising design and designs the model of advertising design teaching system under the concept of curriculum ideology and political education. Secondly, on the basis of the classical division clustering algorithm, by adding a fuzzy set, fuzzy clustering and fuzzy center of mass and explaining the importance of distance, a weighted fuzzy clustering teaching quality evaluation model based on big data analysis is designed. In the end, the teaching model was employed to evaluate the clustering of teaching quality by accumulating educational data. The percentage of excellent teaching quality in the clustering results is about 0.35, the smallest average percentage is failing and average, which are 0.12 and 0.11, respectively, and the highest average weight is the teacher’s professional knowledge and the student’s learning achievement, which are 0.12 and 0.125, respectively.

Generalized Ordered Intuitionistic Fuzzy C-Means Clustering Algorithm Based on PROMETHEE and Intuitionistic Fuzzy C-Means

The problem of ordered clustering in the context of decision-making with multiple criteria has garnered significant interest from researchers in the field of management science and operational research. In real-world scenarios, the datasets often exhibit imprecision or uncertainty, which can lead to suboptimal ordered-clustering outcomes. However, the intuitionistic fuzzy c-means (IFCM) clustering algorithm enhances the accuracy and effectiveness of decision-making processes by effectively handling uncertain dataset information for clustering. Therefore, we propose a new clustering algorithm, called the generalized ordered intuitionistic fuzzy c-means (G-OIFCM), based on PROMETHEE and the IFCM clustering algorithm. Different from the classical IFCM clustering algorithm, we use positive flow ( φ + s i ∈ 0 , 1 ) and negative flow ( φ − s i ∈ 0 , 1 ) of PROMETHEE to generate ordered clusters within the intuitionistic environment. We define a new objective function based on the positive and negative flow of the PROMETHEE and IFCM clustering algorithm, whose properties are mathematically justified in terms of convergence and optimization. The performance of the proposed algorithm is evaluated using two different real-world datasets to assess both the ordered clustering and the quality of partitioning. To demonstrate the effectiveness of G-OIFCM, a comparison is conducted with three other algorithms: fuzzy c-means (FCM), ordered fuzzy c-means (OFCM), and an adaptive generalized intuitionistic fuzzy c-means (G-IFCM). The results demonstrate the effectiveness of G-OIFCM in enhancing optimal ordered clustering and utility when dealing with uncertainty in datasets.

Quantum and Quantum-Inspired Stereographic K Nearest-Neighbour Clustering.

Nearest-neighbour clustering is a simple yet powerful machine learning algorithm that finds natural application in the decoding of signals in classical optical-fibre communication systems. Quantum k-means clustering promises a speed-up over the classical k-means algorithm; however, it has been shown to not currently provide this speed-up for decoding optical-fibre signals due to the embedding of classical data, which introduces inaccuracies and slowdowns. Although still not achieving an exponential speed-up for NISQ implementations, this work proposes the generalised inverse stereographic projection as an improved embedding into the Bloch sphere for quantum distance estimation in k-nearest-neighbour clustering, which allows us to get closer to the classical performance. We also use the generalised inverse stereographic projection to develop an analogous classical clustering algorithm and benchmark its accuracy, runtime and convergence for decoding real-world experimental optical-fibre communication data. This proposed 'quantum-inspired' algorithm provides an improvement in both the accuracy and convergence rate with respect to the k-means algorithm. Hence, this work presents two main contributions. Firstly, we propose the general inverse stereographic projection into the Bloch sphere as a better embedding for quantum machine learning algorithms; here, we use the problem of clustering quadrature amplitude modulated optical-fibre signals as an example. Secondly, as a purely classical contribution inspired by the first contribution, we propose and benchmark the use of the general inverse stereographic projection and spherical centroid for clustering optical-fibre signals, showing that optimizing the radius yields a consistent improvement in accuracy and convergence rate.

A shrinking synchronization clustering algorithm based on a linear weighted Vicsek model

The purpose of clustering is to identify distributions and patterns within unlabelled datasets. Since the proposal of the original synchronization clustering (SynC) algorithm in 2010, synchronization clustering has become a significant research direction. This paper proposes a shrinking synchronization clustering (SSynC) algorithm utilizing a linear weighted Vicsek model. SSynC algorithm is developed from SynC algorithm and a more effective synchronization clustering (ESynC) algorithm. Through analysis and comparison, we find that SSynC algorithm demonstrates superior synchronization effect compared to SynC algorithm, which is based on an extensive Kuramoto model. Additionally, it exhibits similar effect to ESynC algorithm, based on a linear version of Vicsek model. In the simulations, a comparison is conducted between several synchronization clustering algorithms and classical clustering algorithms. Through experiments using some artificial datasets, eight real datasets and three picture datasets, we observe that compared to SynC algorithm, SSynC algorithm not only achieves a better local synchronization effect but also requires fewer iterations and incurs lower time costs. Furthermore, when compared to ESynC algorithm, SSynC algorithm obtains reduced time costs while achieving nearly the same local synchronization effect and the same number of iterations. Extensive comparison experiments with some class clustering algorithms demonstrate the effectiveness of SSynC algorithm.

IDeLUCS: a deep learning interactive tool for alignment-free clustering of DNA sequences.

We present an interactive Deep Learning-based software tool for Unsupervised Clustering of DNA Sequences (iDeLUCS), that detects genomic signatures and uses them to cluster DNA sequences, without the need for sequence alignment or taxonomic identifiers. iDeLUCS is scalable and user-friendly: its graphical user interface, with support for hardware acceleration, allows the practitioner to fine-tune the different hyper-parameters involved in the training process without requiring extensive knowledge of deep learning. The performance of iDeLUCS was evaluated on a diverse set of datasets: several real genomic datasets from organisms in kingdoms Animalia, Protista, Fungi, Bacteria, and Archaea, three datasets of viral genomes, a dataset of simulated metagenomic reads from microbial genomes, and multiple datasets of synthetic DNA sequences. The performance of iDeLUCS was compared to that of two classical clustering algorithms (k-means++ and GMM) and two clustering algorithms specialized in DNA sequences (MeShClust v3.0 and DeLUCS), using both intrinsic cluster evaluation metrics and external evaluation metrics. In terms of unsupervised clustering accuracy, iDeLUCS outperforms the two classical algorithms by an average of ∼20%, and the two specialized algorithms by an average of ∼12%, on the datasets of real DNA sequences analyzed. Overall, our results indicate that iDeLUCS is a robust clustering method suitable for the clustering of large and diverse datasets of unlabeled DNA sequences. iDeLUCS is available at https://github.com/Kari-Genomics-Lab/iDeLUCS under the terms of the MIT licence.

An Efficient Algorithm for Distance-based Structural Graph Clustering

Structural graph clustering (SCAN) is a classic graph clustering algorithm. In SCAN, a key step is to compute the structural similarity between vertices according to the overlap ratio of one-hop neighborhoods. Given two vertices u and v, existing studies only consider the case when u and v are neighbors. However, the structural similarity between non-neighboring vertices in SCAN is always zero, and using only one-hop neighbors on weighted graphs discards the weights on each edge. Both may not reflect the true closeness of two vertices and may fail to return high-quality clustering results. To tackle this issue, we define and study the distance-based structural graph clustering problem. Given a distance threshold d and two vertices u and v, the structural similarity between u and v is defined as the ratio of their respective neighbors within a distance of no more than d. We show that the newly defined distance-based SCAN achieves better clustering results compared to the vanilla version of SCAN. However, the new definition brings challenges in the computation of final clustering results. To tackle this efficiency issue, we propose DistanceSCAN, an efficient approximate algorithm for solving the distance-based SCAN problem. The main idea of DistanceSCAN is to use all-distances bottom-k sketches (ADS) to speed up the computation of similarities. Given the ADS, we can derive the similarity between two vertices with a bounded cost of O(k). However, to ensure that the estimated similarity has an approximation guarantee, the value of k still needs to be set to as large as thousands. This brings high computational costs when computing the similarities between neighboring vertices. To tackle this issue, we further construct histograms to prune the structural similarity computations of vertices pairs. Extensive experiments on real datasets validate the effectiveness and efficiency of DistanceSCAN.

A Novel Intrusion Detection System for Internet of Healthcare Things Based on Deep Subclasses Dispersion Information

Despite the significant benefits that the Internet of Healthcare Things (IoHT) offered to the medical sector, there are concerns and risks regarding these systems which can delay their wide deployment as they handle sensitive and often life-critical medical information. In addition to the IoHT concerns and risks, there are security constraints which include hardware, software and network constraints that pose a security challenge to these systems. Therefore, security measures need to be deployed that can overcome the concerns, mitigate the risks and meet the constraints of the IoHT. For these reasons, a subclasses intrusion detection system for the IoHT is proposed in this research work based on a novel variation of the standard One-class Support Vector Machine (OSVM), namely Deep Subclass Dispersion One-class Support Vector Machine (Deep SDOSVM), which considers subclasses in the target class, i.e. normal class, in order to minimize the data dispersion within and between subclasses, thereby improving the discriminative power and classification performance of the intrusion detection system. A deep clustering model is used for subclasses generation in the proposed Deep SDOSVM approach, namely the Dynamic Autoencoder Model (DynAE), to overcome the drawbacks of the classical clustering algorithms and further enhance the classification performance of the intrusion detection system. The proposed deep clustering subclasses intrusion detection system was evaluated on the real-world TON_IoT dataset and compared to other state-of-the-art one-class classifiers. Experimentation results have shown that the proposed approach outperformed the other relevant one-class classifiers for network intrusion detection.

Molecular cluster mining of high-grade serous ovarian cancer via multi-omics data analysis aids precise medicine.

HGSOC is a kind of gynecological cancer with high mortality and strong heterogeneity. The study used multi-omics and multiple algorithms to identify novel molecular subtypes, which can help patients obtain more personalized treatments. Firstly, the consensus clustering result was obtained using a consensus ensemble of ten classical clustering algorithms, based on mRNA, lncRNA, DNA methylation, and mutation data. The difference in signaling pathways was evaluated using the single-sample gene set enrichment analysis (ssGSEA). Meanwhile, the relationship between genetic alteration, response to immunotherapy, drug sensitivity, prognosis, and subtypes was further analyzed. Finally, the reliability of the new subtype was verified in three external datasets. Three molecular subtypes were identified. Immune desert subtype (CS1) had little enrichment in the immune microenvironment and metabolic pathways. Immune/non-stromal subtype (CS2) was enriched in the immune microenvironment and metabolism of polyamines. Immune/stromal subtype (CS3) not only enriched anti-tumor immune microenvironment characteristics but also enriched pro-tumor stroma characteristics, glycosaminoglycan metabolism, and sphingolipid metabolism. The CS2 had the best overall survival and the highest response rate to immunotherapy. The CS3 had the worst prognosis and the lowest response rate to immunotherapy but was more sensitive to PARP and VEGFR molecular-targeted therapy. The similar differences among three subtypes were successfully validated in three external cohorts. We used ten clustering algorithms to comprehensively analyze four types of omics data, identified three biologically significant subtypes of HGSOC patients, and provided personalized treatment recommendations for each subtype. Our findings provided novel views into the HGSOC subtypes and could provide potential clinical treatment strategies.

Loading and Unloading Points Identification Based on Freight Trajectory Big Data and Clustering Method

Based on the GPS trajectory data of a freight enterprise in Dalian, China, this paper studies the identification of loading and unloading points by a clustering algorithm. Firstly, by analysing the characteristics of freight loading and unloading behaviour, combined with the spatial and temporal distribution characteristics of truck GPS trajectory data, three characteristic variables of the number of trucks passing through a certain place, the average speed of trucks and the average stay time of trucks in the place are extracted. Then, the clustering algorithm and visual analysis are used to obtain the target cluster, and the POI language of the geographic information is obtained according to the points in the target cluster. The meaning information is crawled to accurately identify the result of the freight loading point. Finally, two classical clustering algorithms, K-means and GMM, are evaluated and compared. The results show that the identification method designed in this paper finally identifies 2,320 freight loading and unloading points from 11,406,000 trajectory data, which can realise the accurate extraction of freight loading and unloading points.

A distributed PageRank-based dynamic partition algorithm to improve distributed energy storages participation in frequency regulation

In high renewable energy sources penetrated power systems, a large number of distributed small-capacity energy storages (ESs) are expected to form virtual power plants (VPPs) and provide comparable frequency regulation capacity to conventional power plants. In this paper, a dynamic and distributed partition method is proposed to cluster ESs into VPPs by selecting the electrical distance and local power demands as the index for partitioning. Utilizing the partition indices, the Personalized PageRank (PPR) scores are used to measure the similarity of all ESs with respect to other ESs in parallel. Then Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is utilized to partition the ESs into different VPPs. Within the VPPs, the secondary frequency control strategy is adopted to achieve the frequency regulation and state of charge (SoC) balancing. Finally, the performance comparison of the proposed partition method with several classic clustering algorithms is conducted to validate the efficiency and the quality of the partition. And numerical studies are performed in an IEEE 39-bus system to verify that the proposed control method can eliminate the frequency deviation and reduce power losses.

Rooted Mahalanobis distance based Gustafson-Kessel fuzzy C-means

Fuzzy c-means (FCM) is a classic unsupervised clustering algorithm in machine learning fields. Euclidean distance is a frequently used distance metric in FCM, but it is only suitable for data with spherical clusters. Therefore, Mahalanobis distance was introduced into Gustafson-Kessel Fuzzy C-Means (GK-FCM) to help improve the performance on data with ellipsoidal clusters. However, GK-FCM and existing Mahalanobis distance based algorithms only focus on squared Mahalanobis distance, because squared Mahalanobis distance based problems are usually convex and easily solvable. But squared Mahalanobis distance is not a perfect metric, because it tends to exaggerate the influence of outliers and lead to unsatisfying results. In this paper, we propose a rooted Mahalanobis distance based GK-FCM model, which has better clustering performance and superior robustness than traditional GK-FCM. Moreover, owing to the introduction of rooted Mahalanobis distance, the optimization of the proposed model becomes non-trivial and it is not realistic to obtain a closed-form solution as that of traditional GK-FCM. In this paper, by making reference to the re-weighted method, we develop a novel iterative converging algorithm to optimize the proposed model. Finally, extensive experiments are conducted on both synthetic and real-world data sets to manifest the superiority of the proposed model.