Proposed Clustering Algorithm Research Articles

SubtStream: Online subtractive stream clustering algorithm

AbstractReal‐time stream data processing has gained high importance with the rapid rise of big data trends in different areas such as social media, finance, business, science, and bioinformatics. Stream data can be characterized as fast, unstable, and big data sets. Due to these properties of stream data, it cannot be processed effectively with traditional algorithms. Just like stream data processing, clustering is also a difficult task. However, researchers have attempted to classify stream data by modifying traditional algorithms or designing new ones. So, in previous studies, incremental methods were used for clustering the stream data. This paper highlights the need to develop an efficient real‐time clustering algorithm for data streams in the presence of concept high drift and an adaptive algorithm for different dimensions. The proposed clustering algorithm, SubtStream, combines decremental (subtractive property) and incremental (additivity property) strategies to overcome the high drift. It also introduces a new dimension‐based approach to adopt the dimension change. We use three radius parameters, Predefined User Parameter, Proactive Adaptive Parameter, and Reactive Adaptive Parameter, to achieve adaptability. The proposed method, SubtStream, showed better performance on synthetic and real data sets.

A Fully Bayesian Inference with Gibbs Sampling for Finite and Infinite Discrete Exponential Mixture Models

ABSTRACT In this paper, we propose clustering algorithms based on finite mixture and infinite mixture models of exponential approximation to the Multinomial Generalized Dirichlet (EMGD), Multinomial Beta-Liouville (EMBL) and Multinomial Shifted-Scaled Dirichlet (EMSSD) with Bayesian inference. The finite mixtures have already shown superior performance in real data sets clustering using the Expectation–Maximization approach. The proposed approaches in this paper are based on a Monte Carlo simulation technique namely Gibbs sampling algorithm including an additional Metropolis–Hastings step, and we utilize exponential family conjugate prior information to construct their posterior relying on Bayesian theory. Furthermore, we also present the infinite models based on Dirichlet processes, which results in clustering algorithms that do not require the specification of the number of mixture components to be given in advance and selects it in a principled manner. The performance of our Bayesian approaches was evaluated in some challenging real-world applications concerning text sentiment analysis, fake news detection, and human face gender recognition.

Clustering Based on Continuous Hopfield Network

Clustering aims to group n data samples into k clusters. In this paper, we reformulate the clustering problem into an integer optimization problem and propose a recurrent neural network with n×k neurons to solve it. We prove the stability and convergence of the proposed recurrent neural network theoretically. Moreover, clustering experiments demonstrate that the proposed clustering algorithm based on the recurrent neural network can achieve the better clustering performance than existing clustering algorithms.

Prediction of students’ employability using clustering algorithm: A hybrid approach

Data Mining is a process of exploring the huge data in search of reliable patterns and methodical relationship among variables. As a result, the findings may be validated through applying the detected patterns to a novel subset of the data. In simple words, Data Mining is referred as extracting the useful information as large dataset and transforming into reliable structure for future use. Data Mining has shown its incredible performance in various fields to a greater extent, out of which, Educational Data Mining (EDM) is one among them. Many researchers have addressed huge number of problems in EDM and applied various techniques to reveal the useful and hidden information that helped in the process of decision making. Students getting employed during and after graduation are one of the important parts of their life. Students, based on their academic performances, are getting employed in companies they deserve. But still, the probability of getting employed is very less in this competitive world. In this paper, a real-time scenario has been chosen for analyzing various factors for getting employed/unemployed. Various clustering and classification techniques have been implemented and their performances are studied. A hybrid approach is presented in this paper that integrates the benefits of particle swarm optimization (PSO) and fuzzy clustering means (FCMs). The results obtained show that the proposed technique helps in obtaining higher accuracy to other clustering techniques. The proposed clustering algorithm PSO-FCM, accuracy is 34.4%, 36.45% and 28.45% higher than the existing method, time complexity shows 45%, 33% and 49% lower than the existing [Formula: see text]-means clustering, Naïve Bayes clustering and SVM clustering algorithms, respectively.

Cluster analysis of PM2.5 pollution in China using the frequent itemset clustering approach.

In recent years, severe air pollution has frequently occurred in China at the regional scale. The clustering method to define joint control regions is an effective approach to address severe regional air pollution. However, current cluster analysis research on the determination of joint control areas relies on the Pearson correlation coefficient as a similarity measure. Due to nonlinearity and outliers in air pollution data, the correlation coefficient cannot accurately reveal the similarity in air quality between different cities. To bridge this gap, we proposed a method to delineate spatial patterns of PM2.5 pollution and regional boundaries of polluted areas using the frequent itemset clustering approach. The frequent itemsets between cities were first mined, and the support values were employed as interestingness metrics to describe the significance of similar variation patterns between cities. Then, the hierarchical clustering method was applied to identify appropriate areas for joint pollution control. The proposed clustering algorithm exhibits the advantages of not requiring model assumptions and a robustness to the outliers, which is a cost-effective approach to define joint control regions. By analysing urban PM2.5 pollution in China from 2015 to 2018, we obtained results demonstrating that the frequent itemset clustering approach can efficiently determine pollution patterns and can effectively identify regional divisions. The clustering approach could facilitate a greater understanding of PM2.5 spatiotemporal aggregation to design joint control measures among areas. The findings and methodology of this research have important implications for the formulation of clean air policies in China.

A robust method for clustering football players with mixed attributes

A robust fuzzy clustering model for mixed data is proposed. For each variable, or attribute, a proper dissimilarity measure is computed and the clustering procedure combines the dissimilarity matrices with weights objectively computed during the optimization process. The weights reflect the relevance of each attribute type in the clustering results. A simulation study and an empirical application to football players data are presented that show the effectiveness of the proposed clustering algorithm in finding clusters that would be hidden unless a multi-attributes approach were used.

Electricity Pattern Analysis by Clustering Domestic Load Profiles Using Discrete Wavelet Transform

Energy demand has grown explosively in recent years, leading to increased attention of energy efficiency (EE) research. Demand response (DR) programs were designed to help power management entities meet energy balance and change end-user electricity usage. Advanced real-time meters (RTM) collect a large amount of fine-granular electric consumption data, which contain valuable information. Understanding the energy consumption patterns for different end users can support demand side management (DSM). This study proposed clustering algorithms to segment consumers and obtain the representative load patterns based on diurnal load profiles. First, the proposed method uses discrete wavelet transform (DWT) to extract features from daily electricity consumption data. Second, the extracted features are reconstructed using a statistical method, combined with Pearson’s correlation coefficient and principal component analysis (PCA) for dimensionality reduction. Lastly, three clustering algorithms are employed to segment daily load curves and select the most appropriate algorithm. We experimented our method on the Manhattan dataset and the results indicated that clustering algorithms, combined with discrete wavelet transform, improve the clustering performance. Additionally, we discussed the clustering result and load pattern analysis of the dataset with respect to the electricity pattern.

An Algebraic Approach to Clustering and Classification with Support Vector Machines

In this note, we propose a novel classification approach by introducing a new clustering method, which is used as an intermediate step to discover the structure of a data set. The proposed clustering algorithm uses similarities and the concept of a clique to obtain clusters, which can be used with different strategies for classification. This approach also reduces the size of the training data set. In this study, we apply support vector machines (SVMs) after obtaining clusters with the proposed clustering algorithm. The proposed clustering algorithm is applied with different strategies for applying SVMs. The results for several real data sets show that the performance is comparable with the standard SVM while reducing the size of the training data set and also the number of support vectors.

Piecemeal Clustering: a Self-Driven Data Clustering Algorithm

Various approaches have been discussed in the literature for the clustering of data, such as partitioning, hierarchical, and machine learning methods. Most of the approaches require some prior knowledge about the clusters, such as their total number. Furthermore, some previous algorithms are not robust enough to process higher-dimensional data or require a large amount of memory for computations. We propose, herein, a data clustering algorithm, Piecemeal Clustering, that successfully clusters data without prior knowledge of the number of clusters. The proposed clustering algorithm uses the similarity and density of the data to identify the number of clusters in the data set and works with both low- and high-dimensional data. We demonstrate the power of the proposed Piecemeal Clustering algorithm with two real-world data sets. It is found that the proposed algorithm outperforms seven other state-of-the-art algorithms on both of these data sets.

A Data-Driven Clustering Algorithm for Residual Data Using Fault Signatures and Expectation Maximization

Clustering is an important tool in data-driven fault diagnosis to make use of unlabeled data. Collecting representative data for fault diagnosis is a difficult task since faults are rare events. In addition, using data collected from the field, e.g., logged operational data and data from different workshops about replaced components, can result in labelling uncertainties. A common approach for fault diagnosis of dynamic systems is to use residual-based features that filter out system dynamics while being sensitive to faults. The use of conventional clustering algorithms is complicated by that the distribution of residual data from one fault class varies for different realizations and system operating conditions. In this work, a clustering algorithm is proposed for residual data that clusters data by estimating fault signatures in residual space. The proposed clustering algorithm can be used on time-series data by clustering batches of data from the same fault scenario instead of clustering data sample-by-sample. The usefulness of the proposed clustering algorithm is illustrated using residual data from different fault scenarios collected from an internal combustion engine test bench.

A novel data cluster algorithm based on linear regression and residual analysis for human resource management

Human resource management has become an important part of enterprise management. How to select high-quality talent and how to allocate corresponding talent to appropriate work has become an increasingly acute problem. Traditional data cluster methods cannot effectively solve the above problem due to the high-dimensional data. Therefore, we propose a novel data cluster algorithm based on linear regression and residual analysis for human resource management. Improved hybrid entropy weight attribute similarity is adopted for measuring the similarity between objects. The proposed local density calculation method based on k-nearest neighbour (KNN) and Parzen window is used to calculate the density of each object. Then, we utilise the linear regression and residual analysis to select the clustering centre points quickly and automatically, which can eliminate the subjectivity of artificial selection. A new clustering centre objective optimisation model is proposed to determine the real clustering centre. Through theoretical analysis and comparative experiments on artificial data sets and real data sets, it shows that the proposed cluster algorithm can overcome the defects of the original algorithms, and achieve better clustering effect and lower computation time than state-of-the-art methods.

A New Density Peak Clustering Algorithm Based on Cluster Fusion Strategy

When the density peak clustering algorithm deals with complex datasets and the problem of multiple density peaks in the same cluster, the subjectively selected cluster centers are not accurate enough, and the allocation of non-cluster centers is prone to joint and several errors. To solve the above problems, we propose a new density peak clustering algorithm based on cluster fusion strategy. First, the algorithm screens out the candidate cluster centers by setting two new thresholds to avoid the influence of noise points and outliers. Second, the remaining data points are allocated according to the density peak clustering algorithm to obtain the initial clusters. Third, considering the structural characteristics and spatial distribution of datasets, the new definitions of boundary points, inter-cluster intersection density and inter-cluster boundary density are provided. To correctly classify the clustering problems with multiple density peaks in the same cluster, a new cluster fusion strategy is proposed, which not only corrects the joint and several errors in the allocation of data points, but also correctly selects the cluster centers. Finally, to test the effectiveness of the proposed clustering algorithm, which is compared with DPC-KNN, DPC, K-means and DBSCAN on nine synthetic datasets and six real datasets. The experimental results demonstrate that the clustering performance of the proposed algorithm outperforms that of other algorithms.

On the stochastic robustness of complete clustering predictability for a first‐order consensus model

AbstractWe study stochastic noise response on the complete cluster predictability of a first‐order stochastic consensus model on the real line with an odd, bounded, Lipschitz, and monotone coupling function. Emergence of multiclusters is one of the novel dynamical features of the deterministic consensus model, and it has been extensively studied in collective dynamics community. Recently, the second author and his collaborators proposed a complete cluster predictability of the first‐order nonlinear consensus model that can be derived from the Cucker–Smale flocking model on the real line, and they verified that the proposed clustering algorithm explicitly determines the number of asymptotic clusters and their cluster velocities only in terms of the initial data, coupling strength and supremum of coupling function. In this work, we show that the same cluster algorithm does work for the complete clustering predictability problem when the deterministic consensus model is stochastically perturbed in a multiplicative way.

Minimum spanning tree‐based cluster analysis: A new algorithm for determining inconsistent edges

AbstractIn recent years, graph‐based data clustering algorithms have become popular as they perform connectivity‐based rather than centroid‐based partitioning. Methods related to minimum spanning tree (MST)‐based data clustering are types of graph‐based algorithms that can recognize arbitrary shapes of clusters by eliminating inconsistent edges from MST graphs. In all MST‐based data clustering algorithms, definition of inconsistent edges is the main problem that needs to be addressed. The longest edges in MST graphs are considered as inconsistent edges under ideal conditions. Nevertheless, outliers often exist in real‐world tasks, which makes the longest edges inaccurate cluster separation indicators. In this paper, we propose a new data clustering algorithm using MST and a critical distance method. The proposed algorithm solves the main issue of MST‐based data clustering, namely identifying and removing inconsistent edges to obtain clusters even in the event that the dataset contains some outliers. It begins by constructing the MST over a given weighted graph based on Euclidean distance and then splits up the graph into clusters by eliminating inconsistent edges using critical distance as a threshold. Integration of the advantages of both MST and critical distance methodology to obtain optimal clusters is the main contribution of this work. The conducted experimental analysis and results using different datasets prove that our proposed clustering algorithm yields better overall performance compared with the most common data clustering algorithms. Taking the Liver and Tumor datasets as an example, the proposed algorithm outperforms all other clustering algorithms with clustering accuracy equal to 0.579 and 0.660, respectively.

Workload forecasting and energy state estimation in cloud data centres: ML-centric approach

Resource management in data centres continues to be a critical problem due to increased infrastructure complexity and dynamic workload conditions. Workload and energy consumption prediction are crucial for efficient resource management decisions in cloud data centres. Existing solutions only consider forecasting the usage of virtual machine resources such as CPU and memory; they do not consider provisioned resources (CPU and memory) and disk, network transmission rates, which significantly affect the energy consumption of the host as well. VM-level energy consumption can be estimated for automated energy management decisions in modern data centres. However, it is not easy to measure energy for VM devices such as CPU, memory, and disk at the software level. In this way, we propose an ML-based model to predict load and energy to aid resource management decisions. For modelling workload predictions, we investigated several distinctive ML algorithms such as Linear Regression (LR), Ridge Regression (RR), ARD Regression (ARDR), ElasticNet (EN) and deep learning (DL) algorithm like Gated Recurrent Unit (GRU). The model’s predictions are measured using standard evaluation metrics like root mean square error (RMSE). We have discovered that GRU has performed very well by accomplishing the most negligible RMSE value for all the workload performances based on experimental results. For energy state estimation, we propose four diverse clustering algorithms, including, semi-supervised affinity propagation based on transfer learning (TSSAP), CLA based on transfer learning (TCLA), kmeans based on transfer learning (TKmeans), P-teda based on transfer learning (TP-teda) to discover similar groups of VMs dependent on features that may influence energy consumption as opposed to estimating it for each VM. The TSSAP has acquired promising clustering accuracy with 87.48% and 53.80% in identifying the VM classes which have been calculated using standard metric such as micro-precision for the chosen workload in compassion to affinity propagation (AP) and the average of other proposed clustering algorithms respectively.

CGFFCM: Cluster-weight and Group-local Feature-weight learning in Fuzzy C-Means clustering algorithm for color image segmentation

The fuzzy c-means (FCM) algorithm is a popular method for data clustering and image segmentation. However, the main problem of this algorithm is that it is very sensitive to the initialization of primary clusters, so it may not perform well in segmenting complex images. Another problem with the FCM is the equal importance of the image features used during the segmentation process, which causes unstable performance on different images. In this paper, we propose an FCM-based color image segmentation approach, termed CGFFCM, applying an automatic cluster weighting scheme to reduce the sensitivity to the initialization, and a group-local feature weighting strategy to better image segmentation. Also, we combine the proposed clustering algorithm with the Imperialist Competitive Algorithm (ICA) to optimize the feature weighting process. In addition, we apply an efficient combination of image features to increase the segmentation quality. The performance of CGFFCM is evaluated and compared with state-of-the-art methods (such as SMKIFC (semi-supervised surrogate-assisted multi-objective kernel intuitionistic fuzzy clustering), and A-PSO-IT2IFCM (alternate particle swarm optimization based adaptive interval type-2 intuitionistic FCM clustering algorithm)) using the Berkeley benchmark dataset. The results obtained by CGFFCM are 95%, 79%, and 91%, in terms of average Accuracy, NMI, and F-score metrics, respectively, which all are better than the competitors. The implementation source code of CGFFCM is made publicly available at https://github.com/Amin-Golzari-Oskouei/CGFFCM.

Clustering analysis and punishment‐driven consensus‐reaching process for probabilistic linguistic large‐group decision‐making with application to car‐sharing platform selection

AbstractProbabilistic linguistic large‐group decision‐making (LGDM) is becoming a hot topic in the field of decision science. In general, clustering and consensus‐reaching are the two important processes to deal with LGDM problems. Traditional clustering methods are mainly based on opinion similarity measures and cannot directly control the number of decision makers (DMs) that can be contained in a cluster. Situations of noncooperative behavior toward consensus may be encountered during the consensus‐reaching process. This paper first proposes a distance‐to‐center clustering algorithm to classify the large group. The most important characteristic of the algorithm is that it is based on similarity measures of probabilistic linguistic information, and specifies the upper and lower limits on the number of DMs in the cluster. Then, a novel weight‐determining method is developed that considers the size, external consensus level, and inner consensus level of a cluster. A punishment‐driven consensus model is designed to manage opinion differences and noncooperative behaviors. Finally, the proposed clustering algorithm and consensus model are applied to a case study of car‐sharing platform selection. The results and comparative analysis reveal the potential application and effectiveness of this study.

Performance evaluation of hierarchical clustering protocols with fuzzy C-means

The longevity of the network and the lack of resources are the main problems within the WSN. Minimizing energy dissipation and optimizing the lifespan of the WSN network are real challenges in the design of WSN routing protocols. Load balanced clustering increases the reliability of the system and enhances coordination between different nodes within the network. WSN is one of the main technologies dedicated to the detection, sensing, and monitoring of physical phenomena of the environment. For illustration, detection, and measurement of vibration, pressure, temperature, and sound. The WSN can be integrated into many domains, like street parking systems, smart roads, and industrial. This paper examines the efficiency of our two proposed clustering algorithms: Fuzzy C-means based hierarchical routing approach for homogeneous WSN (F-LEACH) and fuzzy distributed energy efficient clustering algorithm (F-DEEC) through a detailed comparison of WSN performance parameters such as the instability and stability duration, lifetime of the network, number of cluster heads per round and the number of alive nodes. The fuzzy C-means based on hierarchical routing approach is based on fuzzy C-means and low-energy adaptive clustering hierarchy (LEACH) protocol. The fuzzy distributed energy efficient clustering algorithm is based on fuzzy C-means and design of a distributed energy efficient clustering (DEEC) protocol. The technical capability of each protocol is measured according to the studied parameters.

Enhanced Image Segmentation: Merging Fuzzy K-Means and Fuzzy C-Means Clustering Algorithms for Medical Applications

Separating an image into regions according to some criterion is called image segmentation. This paper presents an algorithm that combines the fuzzy k-means (FKM) and fuzzy c-means (FCM) clustering strategies. The proposed algorithm combines the FKM and the FCM algorithms mathematical features, which is referred to as (CFKCM). The FKM and FCM clustering algorithms are adopted to compare the performance and hence evaluate the proposed clustering algorithm. Tests are conducted, and performance parameters are calculated for validation. The comparison and assessment analysis are based on metrics related to the image clustering process, such as the Segmentation Accuracy (SA), Clustering Fitness (CF), and cluster Validity function (V_pc and V_pe). A dataset of MR images is used by this research for the application, test, and evaluation of the image clustering. The results for clustering backbone MRI images show that CFKCM algorithm is featured by being more effective, and comparatively independent of the noise, where it can process the

A state-of-art optimization method for analyzing the tweets of earthquake-prone region

With the increase in accumulated data and usage of the Internet, social media such as Twitter has become a fundamental tool to access all kinds of information. Therefore, it can be expressed that processing, preparing data, and eliminating unnecessary information on Twitter gains its importance rapidly. In particular, it is very important to analyze the information and make it available in emergencies such as disasters. In the proposed study, an earthquake with the magnitude of Mw = 6.8 on the Richter scale that occurred on January 24, 2020, in Elazig province, Turkey, is analyzed in detail. Tweets under twelve hashtags are clustered separately by utilizing the Social Spider Optimization (SSO) algorithm with some modifications. The sum-of intra-cluster distances (SICD) is utilized to measure the performance of the proposed clustering algorithm. In addition, SICD, which works in a way of assigning a new solution to its nearest node, is used as an integer programming model to be solved with the GUROBI package program on the test data-sets. Optimal results are gathered and compared with the proposed SSO results. In the study, center tweets with optimal results are found by utilizing modified SSO. Moreover, results of the proposed SSO algorithm are compared with the K-means clustering technique which is the most popular clustering technique. The proposed SSO algorithm gives better results. Hereby, the general situation of society after an earthquake is deduced to provide moral and material supports.