Partitional Clustering Research Articles

A short-term power prediction method for wind farm cluster based on the fusion of multi-source spatiotemporal feature information

In recent years, the installed capacity of wind power has rapidly increased. And the wind power prediction is the foundation for ensuring large-scale wind power grid connection. The current short-term prediction methods of wind farm cluster (WFC) are difficult to sufficiently extract spatiotemporal features to achieve high-precision prediction. The article proposes a short-term power prediction method for WFC based on deep attention embedded graph clustering-TimesNet (DAEGC-TimesNet). Firstly, the directed power curves of WFC are proposed to analyze wind data. Then, a graph attention network is constructed based on geographic location and numerical weather prediction (NWP) information to guide clustering algorithms to achieve effective cluster partitioning. Finally, the input of model is constructed based on the feature information from various sub-clusters of WFC and the prediction result is obtained through the TimesNet. The method is applied to WFCs in Inner Mongolia, Jilin province and Yunnan province of China. The result shows that the RMSE reduces 0.0155 and the MAE reduces 0.0156 as well as the coefficient of determination and accuracy rate are highest comparing with comparative algorithms averagely based on above three WFCs. The simulation results are superior to the comparison algorithms, which makes greater contributions to ensure regional power supply.

A two-stage clustering ensemble algorithm applicable to risk assessment of railway signaling faults

Knowledge graph (KG) modeling constructs a connected network of hazard/fault events by drawing on historical reports, helping extract risk factors and propagation features in railway signaling safety. For the huge graph structure of the knowledge system, efficiently and accurately transforming the event information with text for risk level prediction is the crucial topic of this paper. To establish a quantitative risk assessment method based on KG modeling, we innovatively introduce text clustering technology to intelligently divide the entity short text data set in the KG, assign standardized entity names to the cluster partitions, and then complete the calculation and analysis according to the characteristic parameter formula, greatly reducing the labor and time consumption of data annotation and approximate text repetitive processing. Significantly, a two-stage clustering ensemble algorithm (TSCEA) integrating base clusterer optimization and hierarchical clustering prediction is proposed, whose innovative approach lies in describing sample similarity by the co-association matrix, while considering the contribution discrepancy of samples for decision-making. Specifically, Stage One selects and optimizes various base clusterers with different method kernels, while Stage Two emphasizes the contribution discrepancy of the cluster core and the cluster halo to make the final clustering decision, with the ensemble strategy serving as the bridge between these two. The KG established on field investigation is used to complete the experiment and analysis of the novel algorithm. Another clustering ensemble algorithm (based on a direct weighting strategy) and each base clusterer are applied to verify clustering performance, while the evaluation depends on external indicators. The results indicate that the proposed clustering ensemble algorithm outperforms others in various indicators, significantly improving clustering accuracy and robustness.

Overlapping coefficient in network-based semi-supervised clustering

AbstractNetwork-based Semi-Supervised Clustering (NeSSC) is a semi-supervised approach for clustering in the presence of an outcome variable. It uses a classification or regression model on resampled versions of the original data to produce a proximity matrix that indicates the magnitude of the similarity between pairs of observations measured with respect to the outcome. This matrix is transformed into a complex network on which a community detection algorithm is applied to search for underlying community structures which is a partition of the instances into highly homogeneous clusters to be evaluated in terms of the outcome. In this paper, we focus on the case the outcome variable to be used in NeSSC is numeric and propose an alternative selection criterion of the optimal partition based on a measure of overlapping between density curves as well as a penalization criterion which takes accounts for the number of clusters in a candidate partition. Next, we consider the performance of the proposed method for some artificial datasets and for 20 different real datasets and compare NeSSC with the other three popular methods of semi-supervised clustering with a numeric outcome. Results show that NeSSC with the overlapping criterion works particularly well when a reduced number of clusters are scattered localized.

Double-Constraint Fuzzy Clustering Algorithm

Given a set of data objects, the fuzzy c-means (FCM) partitional clustering algorithm is favored due to easy implementation, rapid response, and feasible optimization. However, FCM fails to reflect either the importance degree of the individual data objects or that of the clusters. Numerous variants of FCM have been proposed to address these issues. However, most of them cannot effectively apply the available information on data objects or clusters. In this paper, a double-constraint fuzzy clustering algorithm is proposed to reflect the importance degrees of both individual data objects and clusters. By incorporating double constraints into each data object and cluster, the objective function of FCM is reformulated and its realization equation is mathematically conducted. Consequently, the clustering accuracy of FCM is improved by applying the available information on both data objects and clusters. Especially, the proposed algorithm effectively addresses the limitations inherent in the existing variants of FCM. The experimental results validate the effectiveness, implementation, and robustness of the new fuzzy clustering algorithm.

Clinical Presentation, Pathological Spectrum, and Outcomes of Alcoholic Cirrhosis-Related Immunoglobulin A Nephropathy

BackgroundIgA nephropathy associated with cirrhosis is frequent but often overlooked as largely considered silent. Until now, little has been known about their presentation and outcomes. MethodsWe conducted a retrospective multicenter study on patients with kidney biopsy-proven cirrhosis-related IgA nephropathy (cirrhosis-IgAN), diagnosed between 2009 and 2022. We mixed them up with 83 primary IgA nephropathy (pIgAN) diagnosed during the same period, using a partitioning clustering approach, to determine common clinicopathological profiles. ResultsAll the 46 patients with cirrhosis-related IgA nephropathy had an excessive alcoholic consumption. Clinical presentation was severe with acute kidney injury (AKI) in 79%, alternative causes of AKI was found in 62% of cases. Three clinicopathological clusters were identified: one represented chronic involvement, the second could be assimilated to mild disease, and the third corresponded to a MPGN pattern and was associated with heavy proteinuria and intrinsic AKI (without alternative causes). While the two first were equally distributed between pIgAN and cirrhosis-IgAN, the third was more frequent in patients with cirrhosis. The cumulative mortality rate in cirrhosis-IgAN was 26% and 46% at 1-year and 3-years. Steroid exposure and moderate/severe AKI were associated with higher mortality and steroid exposure was associated with the occurrence of severe infection. ConclusionOur results suggest that high AKI incidence is related to extrinsic causes in most cases but can also be driven by IgA-dominant MPGN in a subset of patients. Steroid use was associated with infectious disease and mortality. Further studies are needed to clarify the place of immunosuppressive treatment in cirrhosis-IgAN patients.

Large-scale distributed photovoltaic cluster partition method based on SLM algorithm

The traditional distribution network with centralized control faces the problems of communication delay, large amount of calculation and too many control devices. This research offers a distributed power cluster division technique based on Smart Local Moving (SLM) to overcome the problem of challenging regulation of distributed power supply in modern distribution networks. First of all, on the basis of the modularity division standard, introduce the comprehensive cluster division index of intra-group reactive power and active matching degree index, so as to ensure the strength of the cluster structure on the basis of making it have a certain degree of voltage regulation capability. Secondly, the cluster division scheme of distributed power supply is formed based on SLM algorithm. Lastly, the feasibility and efficacy of the suggested method are validated using IEEE33 cluster partitioning.

Demand Response Strategy Based on the Multi-Agent System and Multiple-Load Participation

In order to improve the utilization of user-side power resources in the distribution network and promote energy conservation, this paper designs a distributed system suitable for power demand response (DR), considering multi-agent system (MAS) technology and consistency algorithms. Due to the frequent changes in the power system structure caused by changes in the load of a large number of users, this paper proposes using cluster partitioning indicators as communication weights between agents, enabling agents to utilize the distribution network for collaborative optimization. In order to achieve the integration of multiple load-side power resources and improve the refinement level of demand-side management (DSM), two types of agents with load aggregator (LA) functions are provided, which adopt the demand response strategies of Time-of-Use (TOU) or Direct Load Control (DLC) and model the uncertainty of individual device states using Monte Carlo method, so that the two typical flexible loads can achieve the target load-reduction requirements under the MAS framework. The research results demonstrate that this method achieves complementary advantages of the two types of loads participating in DR on a time scale, reducing the costs of power companies and saving customers’ electricity bills while peak shaving.

Determining The Optimal Number of K-Means Clusters Using The Calinski Harabasz Index and Krzanowski and Lai Index Methods for Groupsing Flood Prone Areas In North Sumatra

The k-means algorithm is a partitional clustering method. K-Means has several advantages, including being easy to implement, having a high level of convergence and producing denser clusters. Meanwhile, the drawback is that it is difficult to determine the optimal number of clusters. The K-Means method will be used to solve problems in areas prone to flood disasters in North Sumatra. This research aims to find the optimal number of clusters with the Calinski Harabasz Index and Krzanowski And Lai Index based on the Cluster Tightness Measure (CTM) value. There are eleven variables used in this research. Based on the research results, it was concluded that the CTM CH result of 0.376 was smaller than the CTM KL of 0.7843. So it can be said that determining the optimal number of clusters using CH with k = 6 is better than KL with k = 2.

Intelligent mapping of geochemical anomalies: Adaptation of DBSCAN and mean-shift clustering approaches

Cluster analysis can be used to organize samples and generate ideas regarding the multivariate geochemistry of given dataset. Traditional clustering techniques have the drawbacks of high computational complexity and poor adaptability to big data. Hence, there has recently been much focus on creating better clustering algorithms. Although many clustering algorithms have been applied, and some produce notable clustering results, the performance efficiency of algorithms is often highly dependent on the values the user chooses for the parameters. Currently, density-based spatial clustering of applications with noise clustering (DBSCAN) is widely utilized in image processing, bioinformatics, and social network analysis owing to its ability to detect clusters of various shapes. Even though partitional clustering techniques may be effective when the number of clusters K is known in advance, they cannot implement non-convex clustering and rapidly return to a local optimum. This study proposes the concept of DBSCAN clustering for stream sediment geochemical data. In this respect, the geochemical data collected from Varcheh district, SW Iran, were processed using the clr transformation before applying DBSCAN. Then, PCA was used to minimize the dimension of variables and specify the mineralization-related elements. In the following, one of the PCs connected with mineralization (PC2) was chosen for further analysis. DBSCAB, Mean-shift and Fuzzy K-means algorithms were used to monitor the multi-element geochemical anomalies linked to MVT PbZn deposits in the study area. According to Davies-Bouldin and Silhouette as two validation metrics, it can be deduced that the three SCB models are advantageous, however, the model generated by DBSCAN is preferable to the model generated by Mean-shift and Fuzzy K-means.

An integrated framework for UAV-based precision plant protection in complex terrain: the ACHAGA solution for multi-tea fields.

UAV-based plant protection represents an efficient, energy-saving agricultural technology with significant potential to enhance tea production. However, the complex terrain of hilly and mountainous tea fields, coupled with the limited endurance of UAVs, presents substantial challenges for efficient route planning. This study introduces a novel methodological framework for UAV-based precision plant protection across multiple tea fields, addressing the difficulties in planning the shortest routes and optimal flights for UAVs constrained by their endurance. The framework employs a hyperbolic genetic annealing algorithm (ACHAGA) to optimize UAV plant protection routes with the objectives of minimizing flight distance, reducing the number of turns, and enhancing route stability. The method involves two primary steps: cluster partitioning and sortie allocation for multiple tea fields based on UAV range capabilities, followed by refining the UAV's flight path using a combination of hyperbolic genetic and simulated annealing algorithms with an adaptive temperature control mechanism. Simulation experiments and UAV route validation tests confirm the effectiveness of ACHAGA. The algorithm consistently identified optimal solutions within an average of 40 iterations, demonstrating robust global search capabilities and stability. It achieved an average reduction of 45.75 iterations and 1811.93 meters in the optimal route, with lower variation coefficients and extreme deviations across repeated simulations. ACHAGA significantly outperforms these algorithms, GA, GA-ACO, AFSA and BSO, which are also heuristic search strategies, in the multi-tea field route scheduling problem, reducing the optimal routes by 4904.82 m, 926.07 m, 3803.96 m and 800.11 m, respectively. Field tests revealed that ACHAGA reduced actual flight routes by 791.9 meters and 359.6 meters compared to manual and brainstorming-based planning methods, respectively. Additionally, the algorithm reduced flight scheduling distance and the number of turns by 11 compared to manual planning. This study provides a theoretical and technical foundation for managing large-scale tea plantations in challenging landscapes and serves as a reference for UAV precision operation planning in complex environments.

Identifying the partition of port clusters and hierarchical structure for port networks

Identifying the partition of port clusters and hierarchical structure for port networks

A social network user behaviour data recommendation system based on fuzzy partition clustering

A social network user behaviour data recommendation system based on fuzzy partition clustering

A social network user behaviour data recommendation system based on fuzzy partition clustering

A social network user behaviour data recommendation system based on fuzzy partition clustering

EHR Innovations: Shedding Light on Anemia in the Healthcare Paradigm

This study introduces a novel approach to Electronic Health Record (EHR) analysis, extending the use of phenotyping with machine learning (ML) models to enhance the recognition and treatment of anemia. It first examines the healthcare scenario in India and suggests potential improvements through data-driven personalized care. Using the MIMIC-III dataset, the research involves extensive data preprocessing and analysis to uncover key insights into anemia’s prevalence, gender distribution, comorbidities, and Intensive Care Unit (ICU) stays. Partitioning clustering algorithms like K-Means, K-medoids, Fuzzy C-means, and hierarchical clustering algorithms such as Agglomerative Clustering, DIANA, and HDBSCAN are used to identify groups of patients with similar medical profiles. The distance metrics employed are Levenshtein and Euclidean distances combined with TF-IDF Vectorization. The effectiveness of these algorithms is evaluated based on Length of Stay (LoS) estimation, a critical parameter in EHR studies. To predict a new patient’s LoS, the patient is at first classified into an existing cluster, which shows the highest support to the patient’s clinical activities. A decision tree regressor is then trained using data from the selected cluster to predict the new patient’s LoS, significantly improving predictive accuracy and reliability. Notably, the HDBSCAN algorithm, applied to the Tf-Idf Vectorizer object, achieves a 50.82% reduction in Root Mean Squared Error (RMSE) compared to baseline model. The novelty of this study lies in proposing an efficient approach for EHR analysis, specifically for predicting ICU patients’ LoS, and identifying the most effective clustering algorithm to improve healthcare delivery for anemic patients in healthcare scenario of India.

Classification of dementia from spoken speech using feature selection and the bag of acoustic words model

<p>Memory disorders and dementia are a central factor in the decline of functioning and daily activities in older individuals. The workload related to standardized speech tests in clinical settings has led to a growing emphasis on developing automatic machine learning techniques for analyzing naturally spoken speech. This study presented a bag of acoustic words approach for distinguishing dementia patients from control individuals based on audio speech recordings. In this approach, each individual's speech was segmented into voiced periods, and these segments were characterized by acoustic features using the open-source openSMILE library. Word histogram representations were formed from the characterized speech segments of each speaker, which were used for classifying subjects. The formation of word histograms involved a clustering phase where feature vectors were quantized. It is well-known that partitional clustering involves instability in clustering results due to the selection of starting points, which can cause variability in classification outcomes. This study aimed to address instability by utilizing robust K-spatial-medians clustering, efficient K-means$ ++ $ clustering initialization, and selecting the smallest clustering error from repeated clusterings. Additionally, the study employed feature selection based on the Wilcoxon signed-rank test to achieve computational efficiency in the methods. The results showed that it is possible to achieve a consistent 75% classification accuracy using only twenty-five features, both with the external ADReSS 2020 test data and through leave-one-subject-out cross-validation of the entire dataset. The results rank at the top compared to international research, where the same dataset and only acoustic features have been used to diagnose patients.</p>

Equitable cluster partition of graphs with small maximum average degree

Equitable cluster partition of graphs with small maximum average degree

Compound Weakly Supervised Clustering.

Clustering is a fundamental and important step in many image processing tasks, such as face recognition and image segmentation. The performance of clustering can be largely enhanced if relevant weak supervision information is appropriately exploited. To achieve this goal, in this paper, we propose the Compound Weakly Supervised Clustering (CSWC) method. Concretely, CSWC incorporates two types of widely available and easily accessed weak supervision information from the label and feature aspects, respectively. To be specific, at the label level, the pairwise constraints are utilized as a kind of typical weak label supervision information. At the feature level, the partial instances collected from multiple perspectives have internal consistency and they are regarded as weak structure supervision information. To achieve a more confident clustering partition, we learn a unified graph with its similarity matrix to incorporate the above two types of weak supervision. On one hand, this similarity matrix is constructed by self-expression across the partial instances collected from multiple perspectives. On the other hand, the pairwise constraints, i.e., must-links and cannot-links, are considered by formulating a regularizer on the similarity matrix. Finally, the clustering results can be directly obtained according to the learned graph, without performing additional clustering techniques. Besides evaluating CSWC on 7 benchmark datasets, we also apply it to the application of face clustering in video data since it has vast application potentiality. Experimental results demonstrate the effectiveness of our algorithm in both incorporating compound weak supervision and identifying faces in real applications.

ModSoft-HP: Fuzzy Microservices Placement in Kubernetes

The growing popularity of microservices architectures generated the need for tools that orchestrate their deployment in containerized infrastructures, such as Kubernetes. Microservices running in separate containers are packed in pods and placed in virtual machines (nodes). For applications with multiple communicating microservices, the decision of which services should be placed in the same node has a certain impact on both the running time and the operation cost of an application. The default Kubernetes scheduler is not optimal in that case. In this work, the service placement problem is treated as graph clustering. An application is modeled using a graph with nodes and edges representing communicating microservices. Graph clustering partitions the graph into clusters of microservices with high-affinity rates. Then, the microservices of each cluster are placed in the same Kubernetes node. A class of methods resorts to hard clustering (i.e., each microservice is placed in exactly one node). We advocate that graph clustering should be fuzzy to allow high-utilized microservices to run in more than one instance (i.e., pods) in different nodes. ModSoft-HP Scheduler is a custom Kubernetes scheduler that takes scheduling decisions based on the results of the ModSoft fuzzy clustering method followed by heuristic packing (HP). For proof of concept, the workloads of two applications (i.e., an e-commerce application, eShop, and an IoT architecture) are given as input to the default Kubernetes Scheduler, the Bisecting K-means, and the Heuristic First Fit (hard) clustering schedulers and to the ModSoft-HP fuzzy clustering method. The experimental results demonstrate that ModSoft-HP can achieve up to 90% reduction of egress traffic, up to 20% savings in response time, and up to 25% less hosting costs compared to service placement with the default Kubernetes Scheduler in the Google Kubernetes Engine.

Fuzzy logic based prediction of retaining wall stability

In geotechnical engineering, retaining walls are widely employed to solve the problem of supporting horizontal loads occurring between two different soil levels. In the traditional retaining wall design, stability checks continue until a safe design is obtained according to selected wall dimensions and soil properties. This design method is a process that is time-consuming and based on trial and error. In this study, the stability control of the retaining wall, which is a complex engineering design, has been carried out with fuzzy logic methods. Adaptive network-based fuzzy inference systems (ANFISs) including Grid Partition (ANFIS-GP) and Substructive Clustering (ANFIS-SC) have been utilized as fuzzy logic methods. The sliding stability criterion of the cantilever retaining wall has been obtained by performing 1024 retaining wall designs which are created using different wall dimensions. Ninety percent and ten percent of the 1024 sliding safety factor values acquired through numerical analyses were respectively allocated to the training and testing phases. The prediction performances of the methods have been evaluated by considering the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Determination (R²) obtained for the sliding safety factors during the training and testing stages. Upon juxtaposing the actual and anticipated sliding safety factors for a dataset comprising 1024 observations, it has become evident that the ANFIS-SC methodology outperforms the ANFIS-GP approach in terms of predictive accuracy. Furthermore, this analysis culminated in the determination that the application of fuzzy logic methods stands as an efficacious and dependable means for checking the stability criteria of retaining walls.

Elucidating Best Geospatial Estimation Method Applied to Environmental Sciences

The aim of this study is to assess and identify the most suitable geospatial interpolation algorithm for environmental sciences. The research focuses on evaluating six different interpolation methods using annual average PM10 concentrations as a reference dataset. The dataset includes measurements obtained from a target air quality network (scenery 1) and a sub-dataset derived from a partitive clustering technique (scenery 2). By comparing the performance of each interpolation algorithm using various indicators, the study aims to determine the most reliable method. The findings reveal that the kriging method demonstrates the highest performance within environmental sciences, with a spatial similarity of approximately 70% between the two scenery datasets. The performance indicators for the kriging method, including RMSE (root mean square error), MAE (mean absolute error), and MAPE (mean absolute percentage error), are measured at 3.2 µg/m3, 10.2 µg/m3, and 7.3%, respectively.This study addresses the existing gap in scientific knowledge regarding the comparison of geospatial interpolation techniques. The findings provide valuable insights for environmental managers and decision-makers, enabling them to implement effective control and mitigation strategies based on reliable geospatial information and data. In summary, this research evaluates and identifies the most suitable geospatial interpolation algorithm for environmental sciences, with the kriging method emerging as the most reliable option. The study’s findings contribute to the advancement of knowledge in the field and offer practical implications for environmental management and planning.