Data Clustering Method Research Articles

Pengelompokan Hasil Evaluasi Pembelajaran Metode Hafalan Al Qur’an Tawazun Menggunakan Metode K-Means

The method of memorizing the Qur'an Tawazun is a method that maximizes the use of the right brain and left brain, allowing a person to memorize, understand, and believe. Each category has several assessment points as a benchmark for the ability of students, which are used to overcome the level of failure of students in each category of learning the method of memorizing the Qur'an tawazun. The results of the evaluation of the learning of the tahfidz Islamic boarding school Daarul Huffadz Indonesia in 2020 were felt to be less than optimal, because the learning process was carried out simultaneously. This can be seen from the difference in scores that are quite different in each category of assessment. Based on the previous problem, it is necessary to group the results of the evaluation of learning the Qur'an memorization method. The goal is that every student gets maximum treatment and provides convenience for the institution, as well as teaching staff to carry out learning. The purpose of this study is to determine the optimum number of clusters as well as members of each cluster by measuring cluster performance using the Davies Bouldin Index (DBI) method and implementing the K-Means algorithm. The K-Means algorithm is a non-hierarchical data clustering method that is able to group large amounts of data, relatively quickly, and efficiently. This study uses 401 observational data and 12 attributes. From the calculation results, the optimal number of clusters lies in 2 clusters, with a Davies-Bouldin Index (DBI) value of 1.439. There are 26 members of cluster 1, and 375 members of cluster 2.

A Random Matrix Model of Regional Economic Disparity Based on Spatial Panel Data Analysis

This paper adopts a random matrix approach to model spatial panel data and applies the model to the analysis of regional economic disparity. A panel data clustering method based on variable influence and response structural relationship clustering is proposed to define the structural relationship distance, cluster the panel data from the structural relationship so that the data within the class have the same structural relationship, and give the specific algorithm for determining the classification number in the clustering process based on the structural relationship. The methods of clustering panel data with the presence of linear, nonlinear, and functional relationships are considered separately to aggregate and classify the panel data in terms of the structural relationships of parameters. Using the classification modeling approach proposed in this paper, the empirical analysis is better than the estimation effect in the case of no classification; the empirical test data also show that there is a strong cross-sectional correlation between the economic and financial development data of different regions, and the estimation effect of the model with the correlation treatment is better than the estimation effect of the model without the correlation treatment. The study finds that there is a significant imbalance between financial and economic development, the phenomenon of financial exclusion is serious, and the dynamic influence relationship between financial and economic development in relatively less developed regions is different from that between financial development and economic development in relatively developed regions. In addition to the influencing factors, we should improve the efficiency of investment utilization, give better play to the geographical advantages, keep the industrial structure rationalized, develop the special economy according to the local conditions, and give full play to the spatial effect as well as improve the economic development.

Inventory Grouping to Support IT Business Management with the K-Means Algorithm

Inventory control is an important factor in a company that functions to maintain smooth production. Inventory control is the most important activity in the survival of the company. Fulfillment of stocks of goods or products and recording transactions manually. The fluctuating number of requests from consumers resulted in the stock that Gaptech computer had to prepare to become unstable. In addition, the various and many types of products made stock management inaccurate. by extracting data using one of the data mining methods, namely data grouping. In this study grouping sales reports on Gaptech computer, the data is grouped into 2 groups using one of the data grouping algorithms, namely the K-Means Algorithm. The K-Means algorithm is a non-hierarchical data clustering method that divides data into one or more clusters, so that data with the same characteristics are grouped in the same cluster. The grouping results are used to find out which product groups are in great demand and which are less desirable. The results of this study will produce a software that is created using a method of managing incoming transaction data so that it will make it easier for Gaptech computer store owners to manage stock items.

Prediction of Customer Demand for Hands-On Inquiry-Based Learning (HIBL) Product Based on Big Data Clustering Algorithm

The objective was to identify previously unknown groups in a dataset using various techniques. Significant progress has been made in this field in recent years, resulting in the development of novel and promising clustering algorithms. With the constant advancement of big data technology, research on study tours has also become crucial. Clustering can unearth the potential hidden information in large datasets, thereby facilitating more efficient work. Diverse measures have been proposed to quantify similarity, including the Euclidean distance and data space density. As a result, clustering becomes a multi-objective optimization problem. Clustering algorithms are extensively utilized in data preprocessing, data classification, and big data prediction. In this study, we examine clustering methods for big data from a theoretical perspective to comprehend their correlations across a large number of datasets. In addition, we predicted customer demand for research products using fabricated metrics.

Leveraging big data analytics in 5G‐enabled IoT and industrial IoT for the development of sustainable smart cities

AbstractThere has been an exponential growth in the number of low‐cost heterogeneous sensor devices that are connected to the internet in the existing infrastructure of smart cities in the past decade. These sensors and actuator devices are employed in various industries to collect invaluable data that greatly impacts business decisions. State‐of‐the‐art research is being carried out to process the high‐volume data collected in high velocity streams with high variability in order to draw meaningful insights to cater to business needs in various domains. Big data analytics finds diversified opportunities in 5G‐enabled Internet of Things (IoT) and industrial IoT environment to study the data patterns and produce new results which provides big organizations a conducive environment to take informed decisions. This article presents an exhaustive investigation of the various applications and algorithms of the big data analytics in 5G‐driven IoT and industrial IoT systems with a detailed taxonomy of the existing analytical systems and also the challenges specific to the applications in an IoT environment. A holistic understanding on the importance of security and privacy of big data during the development of smart city infrastructure in high velocity streaming process has been explained for a broad understanding of the big data streaming processes. A thorough analysis of important data mining algorithms such as classification, association rule mining, clustering, and prediction methods for big IoT data from the recent literature also has been presented to provide a clear understanding of existing issues in handling big data in 5G‐enabled IoT for the development of sustainable smart city infrastructure.

Variograms for kriging and clustering of spatial functional data with phase variation

Spatial, amplitude and phase variations in spatial functional data are confounded. Conclusions from the popular functional trace-variogram, which quantifies spatial variation, can be misleading when analyzing misaligned functional data with phase variation. To remedy this, we describe a framework that extends amplitude-phase separation methods in functional data to the spatial setting, with a view towards performing clustering and spatial prediction. We propose a decomposition of the trace-variogram into amplitude and phase components, and quantify how spatial correlations between functional observations manifest in their respective amplitude and phase. This enables us to generate separate amplitude and phase clustering methods for spatial functional data, and develop a novel spatial functional interpolant at unobserved locations based on combining separate amplitude and phase predictions. Through simulations and real data analyses, we demonstrate advantages of our approach when compared to standard ones that ignore phase variation, through more accurate predictions and more interpretable clustering results.

An efficient framework for the similarity prediction with query recommendation in E‐learning system

An efficient framework for the similarity prediction with query recommendation in E‐learning system

Self-Supervised Symmetric Nonnegative Matrix Factorization

Symmetric nonnegative matrix factorization (SNMF) has demonstrated to be a powerful method for data clustering. However, SNMF is mathematically formulated as a non-convex optimization problem, making it sensitive to the initialization of variables. Inspired by ensemble clustering that aims to seek a better clustering result from a set of clustering results, we propose self-supervised SNMF (S <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> NMF), which is capable of boosting clustering performance progressively by taking advantage of the sensitivity to initialization characteristic of SNMF, without relying on any additional information. Specifically, we first perform SNMF repeatedly with a random positive matrix for initialization each time, leading to multiple decomposed matrices. Then, we rank the quality of the resulting matrices with adaptively learned weights, from which a new similarity matrix that is expected to be more discriminative is reconstructed for SNMF again. These two steps are iterated until the stopping criterion/maximum number of iterations is achieved. We mathematically formulate S <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> NMF as a constrained optimization problem, and provide an alternative optimization algorithm to solve it with the theoretical convergence guaranteed. Extensive experimental results on 10 commonly used benchmark datasets demonstrate the significant advantage of our S <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> NMF over 14 state-of-the-art methods in terms of 5 quantitative metrics. The source code is publicly available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/jyh-learning/SSSNMF</uri> .

Estimation of fluid permeability of reservoir rocks at great depths using the fuzzy linear regression method

The possibility of using the theory of fuzzy sets is considered. research in assessing the reserves of hard-to-recover deep-immersed hydrocarbons. At the same time, taking into account the complexity of the relationships between individual petrophysical characteristics, on the one hand, and the uncertainty of relevant information, on the other hand, fuzzy logic and flexible computing methods have been found to be more effective. In particular, the data clustering method (Sugeno fuzzy models) with the selection of child functions (membership functions) was tested. In this method, the prediction of the properties of reservoir rocks at great depths is based on a fuzzy linear regression reflecting the interdependence of properties and the natural uncertainty of information. The method was tested on real indicators of the quality of reservoirs of a well-known group of deposits in the Baku archipelago in Azerbaijan. The results of predicting the expected quality indicators of reservoirs at great depths indicate that in the section of the studied fields at depths of more than 4900 m, a decrease in the relative clay content and density of the reservoirs can be expected, but an increase in the permeability for liquid fluids is also possible. Ketwords: deep-seated; hard-to-recover hydrocarbon reserves; fuzzy set theory; reservoir quality prediction.

Fuzzy clustering for multiview data by combining latent information

Multiview data has become very important because it is often possible to obtain multiple representations for the same set of objects. From the perspective of soft partition, this paper proposes a novel fuzzy clustering method for multiview data by combining the latent information or the membership matrices from the classical Fuzzy C-means (FCM) in each view. Considering that multiview data are generated from the same latent subspace, to assist the membership matrix in one view to explore more complementary information from other views, the proposed approach first aligns the set of membership matrices from FCMs in different views to a consensus matrix. To this end, a new objective function of fuzzy clustering is formulated and the optimization method of membership matrices is provided. Then, the optimized latent information or the membership matrix for each view is concatenated into a tensor and the final clustering is derived with the help of tensor decomposition to further exploit high order correlations between different views. To balance the importance of each view, the weighted-view version of the proposed method is also developed. In addition, we analyze the convergence of proposed methods and their computational complexity. Three experimental indices NMI, F-measure and ACC demonstrate that the proposed approach is superior to the latest multiview fuzzy clustering algorithms.

An Unsupervised Clustering Method for Processing Roadside LiDAR Data With Improved Computational Efficiency

In transportation, LiDAR has been primarily used in autonomous vehicles to assist self-driving until recently when people realized it could also be installed at the roadside to support connected vehicles and infrastructure systems. Unlike onboard LiDAR sensors used in autonomous vehicles, roadside applications must perform complete background filtering and clustering as well as tracking real-time traffic movements within the detection zone. This paper presents an unsupervised clustering method for roadside or infrastructure-based LiDAR applications. It first converts 3D LiDAR data points into 2D so that only target points (after background filtering) will be saved in the channel-azimuth 2D structure; then, a method combining the region growing algorithm and counted component labeling is used to perform clustering. Lastly, a merging process is conducted to enhance the connected component labeling method for better outcomes. Experimental studies demonstrate that the proposed method could reach 0.011s per frame (10 Hz sensor rotation frequency) in clustering while maintaining high accuracy.

Conditional functional clustering for longitudinal data with heterogeneous nonlinear patterns

In studies of cognitive aging, it is crucial to distinguish subtypes of longitudinal cognition change while accounting for the effects of given covariates. The longitudinal cognition trajectories and the covariate effects can both be nonlinear with heterogeneous shapes that do not follow a simple parametric form, where flexible functional methods are preferred. However, most functional clustering methods for longitudinal data do not allow controlling for the possible functional effects of covariates. Although traditional mixture-of-experts methods can include covariates and be extended to the functional setting, using nonlinear basis functions, satisfactory parsimonious functional methods required for robust functional coefficient estimation and clustering are still lacking. In this paper we propose a novel latent class functional mixed-effects model in which we assume the covariates have fixed functional effects, and the random curves follow a mixture of Gaussian processes that facilitates a model-based conditional clustering. A transformed penalized B-spline approach is employed for parsimonious modeling and robust model estimation. We propose a new iterative-REML method to choose the penalty parameters in heterogeneous data. The new method is applied to the latest data from the Religious Orders Study and Rush Memory and Aging Project, and four novel subtypes of cognitive changes are identified.

Clustering compositional data using Dirichlet mixture model.

A model-based clustering method for compositional data is explored in this article. Most methods for compositional data analysis require some kind of transformation. The proposed method builds a mixture model using Dirichlet distribution which works with the unit sum constraint. The mixture model uses a hard EM algorithm with some modification to overcome the problem of fast convergence with empty clusters. This work includes a rigorous simulation study to evaluate the performance of the proposed method over varied dimensions, number of clusters, and overlap. The performance of the model is also compared with other popular clustering algorithms often used for compositional data analysis (e.g. KMeans, Gaussian mixture model (GMM) Gaussian Mixture Model with Hard EM (Hard GMM), partition around medoids (PAM), Clustering Large Applications based on Randomized Search (CLARANS), Density-Based Spatial Clustering of Applications with Noise (DBSCAN) etc.) for simulated data as well as two real data problems coming from the business and marketing domain and physical science domain, respectively. The study has shown promising results exploiting different distributional patterns of compositional data.

An Ensemble Clustering Approach (Consensus Clustering) for High-Dimensional Data

Due to the plurality of irrelevant attributes, sparse distribution, and complicated calculations in high-dimensional data, traditional clustering algorithms, such as K-means, do not perform well on high-dimensional data. To address the clustering problem of high-dimensional data, this paper studies an integrated clustering method for high-dimensional data. A method of subspace division based on minimum redundancy is proposed to solve the problem of subspace division of high-dimensional data; subspace division is improved by using the K-means algorithm. Additionally, this method uses mutual information between the characteristic variables of the data to replace the calculation in the K-means algorithm. The distance between the characteristic variables of the data is used to divide the data into subspaces according to the mutual information values between the characteristic variables of the data. To achieve high clustering accuracy and diversity based on clustering requirements, this paper uses a genetic algorithm as the consistency integration function. The fitness function is designed according to the clustering fusion target, and the selection operator is designed according to the maximum number of overlapping elements in the base clustering. The experimental results show that the clustering algorithm proposed in this paper outperforms other methods on most datasets and is an effective clustering integration algorithm. The proposed clustering algorithm is compared with other commonly used clustering fusion algorithms on datasets to prove the advantages of the proposed algorithm.

Adaptive and structured graph learning for semi-supervised clustering

This paper proposes a new method for semi-supervised clustering of data that only contains pairwise relational information. Specifically, our method simultaneously learns two similarity matrices in feature space and label space, in which similarity matrix in feature space learned by adopting adaptive neighbor strategy while another one obtained through tactful label propagation approach. Moreover, the above two learned matrices explore the local structure (i.e., learned from feature space) and global structure (i.e., learned from label space) of data respectively. Furthermore, most of the existing clustering methods do not fully consider the graph structure, they can not achieve the optimal clustering performance. Therefore, our method forcibly divides the data into c clusters by adding a low rank restriction on the graphical Laplacian matrix. Finally, a restriction of alignment between two similarity matrices is imposed and all items are combined into a unified framework, and an iterative optimization strategy is leveraged to solve the proposed model. Experiments in practical data show that our method has achieved brilliant performance compared with some other state-of-the-art methods.

A parameter-free deep embedded clustering method for single-cell RNA-seq data.

Clustering analysis is widely used in single-cell ribonucleic acid (RNA)-sequencing (scRNA-seq) data to discover cell heterogeneity and cell states. While many clustering methods have been developed for scRNA-seq analysis, most of these methods require to provide the number of clusters. However, it is not easy to know the exact number of cell types in advance, and experienced determination is not always reliable. Here, we have developed ADClust, an automatic deep embedding clustering method for scRNA-seq data, which can accurately cluster cells without requiring a predefined number of clusters. Specifically, ADClust first obtains low-dimensional representation through pre-trained autoencoder and uses the representations to cluster cells into initial micro-clusters. The clusters are then compared in between by a statistical test, and similar micro-clusters are merged into larger clusters. According to the clustering, cell representations are updated so that each cell will be pulled toward centers of its assigned cluster and similar clusters, while cells are separated to keep distances between clusters. This is accomplished through jointly optimizing the carefully designed clustering and autoencoder loss functions. This merging process continues until convergence. ADClust was tested on 11 real scRNA-seq datasets and was shown to outperform existing methods in terms of both clustering performance and the accuracy on the number of the determined clusters. More importantly, our model provides high speed and scalability for large datasets.

New seasonal measurement with stability and clustering seasonal patterns: A case study in Japan from 2011 to 2019.

Seasonality of tourism demand witnesses fluctuations over multiple years. The fluctuations and seasonality often cause seasonal pattern changes. This study presents a new seasonal measurement that considers the stability of seasonal patterns. The measurement is based on a seasonal index and expressed interval-valued data, which are a kind of symbolic data. As a case study, the new measurements are calculated from Japanese overnight data from 2011 to 2019 and the results classify the seasonal patterns of 47 prefectures using a hierarchical clustering method for interval-valued data based on Ward's method. The analysis results indicate that there are differences in not only seasonal patterns but also their stability between domestic overnight demand and inbound overnight demand. The analysis procedure suggested in this study could be helpful in organizing numerous other unstable seasonal patterns.

Min–max kurtosis mean distance based k-means initial centroid initialization method for big genomic data clustering

Min–max kurtosis mean distance based k-means initial centroid initialization method for big genomic data clustering

Model-based clustering of high-dimensional longitudinal data via regularization.

We propose a model-based clustering method for high-dimensional longitudinal data via regularization in this paper. This study was motivated by the Trial of Activity in Adolescent Girls (TAAG), which aimed to examine multilevel factors related to the change of physical activity by following up a cohort of 783 girls over 10 years from adolescence to early adulthood. Our goal is to identify the intrinsic grouping of subjects with similar patterns of physical activity trajectories and the most relevant predictors within each group. The previous analyses conducted clustering and variable selection in two steps, while our new method can perform the tasks simultaneously. Within each cluster, a linear mixed-effects model (LMM) is fitted with a doubly penalized likelihood to induce sparsity for parameter estimation and effect selection. The large-sample joint properties are established, allowing the dimensions of both fixed and random effects to increase at an exponential rate of the sample size, with a general class of penalty functions. Assuming subjects are drawn from a Gaussian mixture distribution, model effects and cluster labels are estimated via a coordinate descent algorithm nested inside the Expectation-Maximization (EM) algorithm. Bayesian Information Criterion (BIC) is used to determine the optimal number of clusters and the values of tuning parameters. Our numerical studies show that the new method has satisfactory performance and is able to accommodate complex data with multilevel and/or longitudinaleffects.

Optimized time series data clustering method for NN network

Time series data is a common sort of data that can be found in a variety of fields. Clustering time series data has a wide range of uses and has drawn scholars from several fields. A unique approach for shape-based time series clustering is proposed in this research. Using the complex network principle, it may reduce the bulk of data, enhance efficiency, and not reduce the effects. To begin, a one-nearest neighbour network is constructed using time series items that are comparable. The similarity is measured in this step using the triangle distance. Each node in the neighbour network represents a single time series object, whereas each connection indicates a neighbour relationship between nodes. Second, high-degree nodes are selected and used to cluster. Dynamic time warping distance function and hierarchical clustering algorithm are used in the clustering process. Finally, several studies are carried out using both synthetic and actual data. The findings demonstrate that the suggested approach is both efficient and effective.