Clustering Index Research Articles

Photovoltaic module regional clustering in mainland China and application based on factors influencing field reliability

The field reliability of photovoltaic modules is important to make investment, design, operation and maintenance decisions in PV power generation projects. By minimizing the field reliability differences between the congeneric regions of PV modules, a regional clustering method based on the factors influencing field reliability is proposed, and a model to predict the field reliability or service lifetime by the clustering results is presented. Based on systematic analysis of the regional differences between workload and natural wear factors impacting field reliability, a comprehensive clustering model is constructed, and nine clustering indexes are accurately quantified. The Ward and entropy weight methods are adopted to objectively calculate index and time weights, respectively. The weighted Ward clustering algorithm is applied to cluster regions based on workload and natural wear factors. Two clustering results are integrated through comprehensive clustering to obtain the final result. Finally, this method is applied to the clustering of 31 provincial administrative regions in mainland China, and the result is applied to predict the average annual power degradation amount of the PV modules in the different regions, both of which validate the applicability and effectiveness of the proposed method.

Genetic diversity and population structure of elite drought tolerant bread wheat (Triticum aestivum L.) genotypes

The objective of this study was to assess genetic diversity and population structure of 47 bread wheat genotypes obtained from the International Maize and Wheat Improvement Centre (CIMMYT) using 10 polymorphic simple sequence repeat (SSR) markers. Data was subjected to analysis for generating a dissimilarity matrix by the Jaccard index for clustering by the Neighbour-joining algorithm on DARwin 6.5 software. GenAlex Software was used to analyse the number of detected alleles (Na), number of effective alleles (Ne), observed heterozygosity (Ho), expected heterozygosity (He), genetic distance (GD), genetic identity (GI), gene flow (Nm), fixation index (F), Shannon’s Information Index (I), Analysis of molecular variance (AMOVA) and polymorphic information content (PIC). Results revealed that, at the locus level, marker Xgwm 132 had the highest Na (21), Ne (14.5), Ho (1.0) and He (0.94), while at the population level, Population III had the highest Na (21), Ne (5.59), He (0.83), and I (1.78). The mean PIC recorded was 0.80, and ranged from 0.63 (Wmc 78) to 0.93 (Xgwm 132). AMOVA revealed significant differences in genetic variation allocated within individuals (60%), across different individuals (37%) and across populations (3%) (P < 0.001). Four populations were distinguished based on pedigrees with GD ranging from 0.01 (Populations III and IV) to 0.31 (Populations II and III), while GI ranged from 0.74 (Populations II and III) to 0.99 (Populations III and IV). The selected markers successfully distinguished test genotypes with the most informative marker being Xgwm 132. Populations II and III were most distinct, thus suitable for parental selection and further drought tolerance breeding.

Using seasonal rainfall clusters to explain the interannual variability of the rain belt over the Greater Horn of Africa

AbstractThe seasonal cycle of rainfall over the Greater Horn of Africa (GHA) is dominated by the latitudinal migration and activity of the tropical rain belt (TRB). The TRB exhibits high interannual variability in the GHA and the reasons for the recent dry period in the Long Rains (March–May) are poorly understood. In addition, few studies have addressed the rainfall fluctuations during the Msimu Rains (Dec.–Mar.) in the southern GHA region. Interannual variations of the seasonal cycle of the TRB between 1981 and 2018 were analysed using two statistical indices. The Rainfall Cluster Index (RCI) describes the seasonal cycle as a succession of six characteristic rainfall patterns, while the Seasonal Location Index (SLI) captures the latitudinal location of the TRB. The SLI and RCI depict the full seasonal cycle of the TRB supporting interpretations of the interannual variations and trends. The Msimu Rains are dominated by two clusters with opposite rainfall characteristics between the Congo Basin and Tanzania. The associated anomalies in moisture flux and divergence indicate variations in the location of the TRB originating from an interplay between low‐level air flows from the Atlantic and Indian Oceans and tropical and subtropical teleconnections. The peak period of the Long Rains shows a complex composition of five clusters, which is tightly connected to intraseasonal and interannual variability of latitudinal locations of the TRB. A persistent location of the TRB near the equator, evidenced in a frequent occurrence of a cluster related to an anomalously weak Walker circulation, is associated with wet conditions over East Africa. Dry Long Rains are associated with strong and frequent latitudinal variations of the TRB position with a late onset and intermittent rainfall. These results offer new opportunities to understand recent variability and trends in the GHA region.

Cluster index for estimating thermal poultry stress (gallus gallus domesticus)

Reducing production losses due to thermal stress is a challenge for poultry. The cost of operating HVAC systems is high, and precision in identifying critical situations for further action is necessary. Making decisions based on parameters inherent to animals has been widely studied and the analysis of animal behavior has advantages in terms of the lower cost of this monitoring and the need to not manipulate animals. This work proposes an index to estimate the thermal comfort of poultry birds (gallus gallus domesticus), based on the analysis of the birds' agglomeration behavior, called cluster index. This index is calculated from variables measured through the analysis of digital images, obtained in the filming of poultry groups from a surveillance camera installed on the roof of the farm. The image processing and analysis algorithm was developed on MatLab, considering that the filming was done with a camera installed on top of the birds, it can be applied to any video as long as the poultry have white plumage and are on a bed of dark wood shavings. The cluster index was tested on videos from two different experiments: one with three breeds of broiler breeders (Hybro, Cobb, and Ross) and one with pruning chickens (Nick Chik), bred on wood shavings and with known ambient temperature conditions. The results showed that poultry, whether laying or from heavy breeding, clump less in conditions of high temperature, and that among heavy breeding breeds, there is a significant difference in cluster behavior under conditions of comfort and thermal stress. It is concluded that the proposed cluster index was able to distinguish differences in the agglomeration of birds (gallus gallus domesticus) under different temperature conditions, from the automatic analysis of images and it is suggested that the cluster index associated with a remote monitoring system can be used to determine early thermal stress in poultry bred on shavings.

Technology convergence networks for flexible display application: A comparative analysis of latecomers and leaders

This study investigates whether systematic differences exist between the technology convergence networks of South Korea, a latecomer country in the field of flexible display applications, and Japan, a leading country. Data focusing on materials for flexible display applications were collected from the United States Patent and Trademark Office. Co-classification analysis was performed for the two countries’ technology networks. The processed data were then analyzed using cluster and concentration index analyses. Cluster analysis indicates that South Korea has a homogeneous technology field, whereas inter-field technology convergence is active in Japan. The concentration index analysis shows that South Korea focuses on specific technology fields, while Japan actively engages in distributed technology convergence. Thus, this study contributes to understanding technology convergence by clarifying how technology is connected and integrated by latecomer and leader country in the field of flexible display applications.

OsamorSoft: clustering index for comparison and quality validation in high throughput dataset

The existence of some differences in the results obtained from varying clustering k-means algorithms necessitated the need for a simplified approach in validation of cluster quality obtained. This is partly because of differences in the way the algorithms select their first seed or centroid either randomly, sequentially or some other principles influences which tend to influence the final result outcome. Popular external cluster quality validation and comparison models require the computation of varying clustering indexes such as Rand, Jaccard, Fowlkes and Mallows, Morey and Agresti Adjusted Rand Index (ARIMA) and Hubert and Arabie Adjusted Rand Index (ARIHA). In literature, Hubert and Arabie Adjusted Rand Index (ARIHA) has been adjudged as a good measure of cluster validity. Based on ARIHA as a popular clustering quality index, we developed OsamorSoft which constitutes DNA_Omatrix and OsamorSpreadSheet as a tool for cluster quality validation in high throughput analysis. The proposed method will help to bridge the yawning gap created by lesser number of friendly tools available to externally evaluate the ever-increasing number of clustering algorithms. Our implementation was tested alongside with clusters created with four k-means algorithms using malaria microarray data. Furthermore, our results evolved a compact 4-stage OsamorSpreadSheet statistics that our easy-to-use GUI java and spreadsheet-based tool of OsamorSoft uses for cluster quality comparison. It is recommended that a framework be evolved to facilitate the simplified integration and automation of several other cluster validity indexes for comparative analysis of big data problems.

PENGIDENTIFIKASIAN SEGMENTASI PENGGUNA SISTEM MANAJEMEN PEMBELAJARAN SEBUAH UNIVERSITAS DENGAN METODE TWO-STEP CLUSTERING

This study aims to identify user segmentations of a learning management system at Universitas Negeri Malang using two-step clustering method. Data is collected through an on-line survey which is linked onto the academic information systems. Total number of data are 10.594 responses. This study runs three clustering methods, two-step, k-means, and kohonen clustering, in order to identify the user segmentation. Two-step clustering methods performs better than the other two clustering methods (k-means and kohonen) based on silhouette clustering index. The number of cluster resulted from two-step clustering method are three clusters. The top three of most important factors in identifying the user segmentation are the benefit of the learning management usage in helping to access information, increasing the effectiveness of accomplishing assignments, and students satisfaction on the learning management system services.

Spatial Agglomerations in the Spanish Food Industry: Does Sectorial Disaggregation Matter?

The food industry is the most important industrial activity in Spain in terms of production and employment; its spatial concentration is considerable. However, there is a lack of quantitative research on its agglomerations, especially at high levels of spatial and sectorial disaggregation. With the study presented, this deficit is addressed, using systematic and quantitative methods to examine the spatial agglomeration in the subsectors that the food industry is specialized. Spanish food industry clusters have been identified by applying a top-down quantitative methodological approach—the cluster index—and following a high level of territorial and sectorial disaggregation. Clusters were identified at the three- and four-digit sectorial disaggregation. The number of four-digit-level relevant clusters was higher than three-digit disaggregation. This evidence was verified using nonparametric statistical tests (Wilcoxon signed-rank test). Furthermore, the analysis of four significant three-digit subsectors allows us to advance two alternative explanations for the divergent results at the different levels of the analysis (three digits vs. four digits). This partly explains the potential competitiveness of Spain in the food industry and shows the possibility of establishing policies for the development of clusters.

Dynamic Data Retrieval Using Incremental Clustering and Indexing

The evolution of the Internet and real-time applications has contributed to the growth of massive unstructured data which imposes the increased complexity of efficient retrieval of dynamic data. Extant research uses clustering methods and indexes to speed up the retrieval. However, the quality of clustering methods depends on data representation models where existing models suffer from dimensionality explosion and sparsity problems. As documents evolve, index reconstruction from scratch is expensive. In this work, compact vectors of documents generated by the Doc2Vec model are used to cluster the documents and the indexes are incrementally updated with less complexity using the diff method. The probabilistic ranking scheme BM25+ is used to improve the quality of retrieval for user queries. The experimental analysis demonstrates that the proposed system significantly improves the clustering performance and reduces retrieval time to obtain top-k results.

Visualization Analysis of Seismic Facies Based on Deep Embedded SOM

As a classical visualization tool for seismic facies analysis, the clustering process of self-organizing map (SOM) is generally divided into two stages: feature extraction and clustering. However, when the horizon pickings are ambiguous and waveforms are chaotic, the structural feature extraction cannot correspond well to the topological structure of SOM. To improve the performance of classification, we propose a one-stage method of deep embedded SOM (DESOM) for seismic facies visualization analysis, which means that the extracting feature representation and clustering are completed simultaneously. Moreover, the stability of the DESOM results can be improved by adding sparse constraints, and thus the visualization results can be displayed in more detail. In the experiment, by comparing with the external indexes of clustering and the hierarchical results of prototype categories, the superiorities of the DESOM and sparse DESOM (SDESOM) methods are verified based on a geophysical model. In the field data application, these methods are combined with the Hue, Saturation, Value (HSV) color mapping technology to display the geological structure information of the target horizon. According to the law of the correlation of facies and the internal cluster indexes, it approves that the DESOM method can improve the continuity of channels, and the SDESOM method can obtain more detailed information of seismic facies distribution.

Model-Based Sensitivity Analysis of EMG Clustering Index With Respect to Motor Unit Properties: Investigating Post-Stroke FDI Muscle.

The objective of this study is to explore the diagnostic decision and sensitivity of the surface electromyogram (EMG) clustering index (CI) with respect to post-stroke motor unit (MU) alterations through a simulation approach by the existing motor neuron pool model and surface EMG model. In the simulation analysis, three patterns of diagnostic decisions were presented in 24 groups representing eight types in three degrees of MU alterations. Specifically, the CI decision exhibited an abnormally increased pattern for five types, an abnormally decreased pattern for two types, and an invariant pattern for one type. Furthermore, the CI diagnostic decision was found to be highly sensitive to three types because a 50% degree of alteration in these types resulted in a distinct deviation of 2.5 in the CI Z-score. The mixed CI patterns were confirmed in experimental data collected from the paretic muscles of 14 subjects with stroke, as compared to the healthy muscles of 10 control subjects. Given the simulation results as a guideline, the CI diagnostic decision could be interpreted from general neural or muscular changes into specific MU changes (in eight types). This can further promote clinical applications of the convenient surface EMG tool in examining and monitoring paretic muscle changes toward customized stroke rehabilitation.

Genomic signal processing of microarrays for cancer gene expression and identification using cluster-fuzzy adaptive networking

Genomic signal processing (GSP) is a functioning exploration area of recent times and a settled technique of digital signal processing for gathering information from genomic sequences. The recognition and identification of biological signals and analysis of sequences are the fundamental objectives of using GSP. Microarray data are typically used in GSP; microarray study decides genes that cause a specific disease and helps in anticipating and diagnosing a disease, and characterization of diseases. Microarray information is incredible innovation where information handled to an enormous number with plenty of genes. Recent research works show that microarray handling will be helpful for the classification of cancer genes. Different machine learning and artificial intelligence techniques are likewise used to distinguish the tumours and cancer cells. In this examination, the genomic signal processing is carried out utilizing cluster-fuzzy adaptive networking techniques. The major purpose of this research work is to evaluate the microarray data sets for recognizing the cancer genes. The microarray data set is generated using leukaemia, colon, prostate, breast cancer and lymphoma. Initially, the noise in the microarray is filtered and smoothened by utilizing a Kalman filter followed by an optimal clustering technique such as grid density-based clustering that is applied for clustering the microarray data sets. The clustered data of microarray are classified by adaptive neuro fuzzy interference system (ANFIS) for gene sequencing process of cancer identification. The adaptive network systems are developed based on autonomous networking concepts to change the static system into a dynamic. The efficiency of clustering is evaluated in terms of cluster indexes namely partition entropy, partition coefficient, Xie and Beni. The presented ANFIS is assessed in terms of precision, accuracy, recall, sensitivity, F-score and specificity. The proposed initiated methodology is mathematically designed and executed in the MATLAB platform and run for various test runs. During the implementation, the performance of cluster and classification efficiency of proposed techniques are compared with the existing strategies like fuzzy c-means with ANN and density-based clustering with ANN, respectively. Ultimately, the performance outcomes demonstrated that the proposed method can provide effective and optimal classification and identification of microarray cancer genes through genomic signal processing than the conventional methods, respectively.

Form factor analysis of the surface electromyographic interference pattern.

In contrast to needle electromyography (EMG), surface EMG recordings are painless. It is of interest to develop methods to analyze surface EMG for diagnostic purposes. Surface EMG interference pattern (SIP) recordings from the abductor pollicis brevis muscle of healthy subjects and subjects with amyotrophic lateral sclerosis (ALS) were analyzed by measuring root-mean-square (RMS) voltage, mean rectified voltage, form factor (FF), and the clustering index (CI). The FF vs SIP area plot was used for analysis. The SIP FF was increased and abnormal in ALS subjects, especially when SIP area was less than 200 mVms. Power regression showed a faster FF decline with SIP area in ALS patients than in healthy subjects. The CI and FF showed a strong correlation. FF is easy to calculate and demonstrates abnormalities in ALS patients.

A robust density peaks clustering algorithm with density-sensitive similarity

Density peaks clustering (DPC) algorithm is proposed to identify the cluster centers quickly by drawing a decision-graph without any prior knowledge. Meanwhile, DPC obtains arbitrary clusters with fewer parameters and no iteration. However, DPC has some shortcomings to be addressed before it is widely applied. Firstly, DPC is not suitable for manifold datasets because these datasets have multiple density peaks in one cluster. Secondly, the cut-off distance parameter has a great influence on the algorithm, especially on small-scale datasets. Thirdly, the method of decision-graph will cause uncertain cluster centers, which leads to wrong clustering. To address these issues, we propose a robust density peaks clustering algorithm with density-sensitive similarity (RDPC-DSS) to find accurate cluster centers on the manifold datasets. With density-sensitive similarity, the influence of the parameters on the clustering results is reduced. In addition, a novel density clustering index (DCI) instead of the decision-graph is designed to automatically determine the number of cluster centers. Extensive experimental results show that RDPC-DSS outperforms DPC and other state-of-the-art algorithms on the manifold datasets.

Role of Cluster Validity Indices in Delineation of Precipitation Regions

The delineation of precipitation regions is to identify homogeneous zones in which the characteristics of the process are statistically similar. The regionalization process has three main components: (i) delineation of regions using clustering algorithms, (ii) determining the optimal number of regions using cluster validity indices (CVIs), and (iii) validation of regions for homogeneity using L-moments ratio test. The identification of the optimal number of clusters will significantly affect the homogeneity of the regions. The objective of this study is to investigate the performance of the various CVIs in identifying the optimal number of clusters, which maximizes the homogeneity of the precipitation regions. The k-means clustering algorithm is adopted to delineate the regions using location-based attributes for two large areas from Canada, namely, the Prairies and the Great Lakes-St Lawrence lowlands (GL-SL) region. The seasonal precipitation data for 55 years (1951–2005) is derived using high-resolution ANUSPLIN gridded point data for Canada. The results indicate that the optimal number of clusters and the regional homogeneity depends on the CVI adopted. Among 42 cluster indices considered, 15 of them outperform in identifying the homogeneous precipitation regions. The Dunn, D e t _ r a t i o and Trace( W − 1 B ) indices found to be the best for all seasons in both the regions.

A Novel Clustering Index to Find Optimal Clusters Size With Application to Segmentation of Energy Consumers

Increased deployment of residential smart meters has made it possible to record energy consumption data on short intervals. These data, if used efficiently, carry valuable information for managing power demand and increasing energy consumption efficiency. An efficient way to analyze these data is to first identify clusters of energy consumers, and then focus on analyzing these clusters. However deciding on the optimal number of clusters is a challenging task. In this article, we propose a clustering index that effectively finds the optimal number of clusters. The proposed index is an entropy-based measure that is obtained from eigenvalue analysis of the correlation matrix of time series of consumption data. A genetic algorithm based feature selection is used to reduce the number of features, which are then fed into clustering algorithms. We apply the proposed clustering index on two ground truth synthetic data sets and two real world energy consumption data set. The numerical simulations reveal the effectiveness of the proposed method and its superiority to a number of existing clustering indices.

Clustering algorithm based on analytic hierarchy process in LTE-V2V

As a hot area of urban transportation, the development of internet of vehicles and related technologies has received widespread attention. This paper proposes an analytic hierarchy process clustering algorithm (AHP-CA) based on the analytic hierarchy process in the LTE vehicle self-organizing network in the scenario of urban two-direction road intersections. The AHP-CA algorithm introduces the average distance between vehicles, normalizes the average relative speed, and uses the average relative motion direction as the vehicle eigenvalue for the clustering index. The analytic hierarchy process (Analytic Hierarchy Process, AHP) is used to determine the weights of different vehicle feature values, and the appropriate nodes are selected as cluster heads by combining the weights. The simulation results verify the stability of the cluster head under different vehicle speeds and communication distances.

Bearing fault diagnosis based on teager energy entropy and mean-shift fuzzy C-means

Feature extraction and fault recognition of vibration signals are two important parts of bearing fault diagnosis. In this article, a fault diagnosis method based on teager energy entropy of each wavelet subband and improved fuzzy C-means is proposed. First, bearing vibration signal is decomposed into wavelet packet and normalized teager energy entropy feature matrix is constructed as clustering index. Principal component analysis is applied to the high-dimensional teager energy entropy feature matrix, and the principal components are determined by cumulative contribution rate to construct feature vectors. Then, the mean-shift method is used to search for the high probability density region of principal components so as to determine the cluster number and cluster center. Finally, fuzzy C-means is used to update the clustering center and membership value, and confirm the optimal clustering center and the type of clustering. Through simulated and experimental analysis, the proposed method has two advantages. The feature vector constructed by this method has better specificity than wavelet energy entropy. The initial clustering center of fuzzy C-means is confirmed by the mean-shift method, which can improve the clustering performance of fuzzy C-means and solve the misclassification without preknowing the number of categories.

3D soil void space lacunarity as an index of degradation after land use change

In this work, lacunarity analysis is performed on soil pores segmented by the pure voxel extraction method from soil tomography images. The conversion of forest to sugarcane plantation was found to result in higher sugarcane soil pore lacunarity than that of native forest soil, while the porosity was found to be lower. More precisely, this study shows that native forest has more porous soil with a more uniform spatial distribution of pores, while sugarcane soil has lower porosity and a more heterogeneous pore distribution. Moreover, validation through multivariate statistics demonstrates that lacunarity can be considered a relevant index of clustering and can explain the variability among soils under different land use systems. While porosity by itself represents a fundamental concept for quantification of the impact of land use change, the current findings demonstrate that the spatial distribution of pores also plays an important role and that pore lacunarity can be adopted as a complementary tool in studies directed at quantifying the effect of human intervention on soils.

Evaluation and Analysis of Electronic Information Industry Clustering Level Based on Soft Subspace Clustering Algorithm

Based on the soft subspace clustering, this paper constructs an industry clustering index which can be used to test the significance, and then calculates and analyzes the clustering level of China’s electronic information industry. The results show that the clustering phenomenon of China’s electronic information industry is very obvious, and the clustering index shows an inverted U-shaped trend of rapid rise and slow decline in the whole research range, In terms of the distribution of the gathering areas, they are mainly concentrated in the eastern coastal areas such as Guangdong, Jiangsu and Shanghai, and the concentration degree is constantly increasing.