Fuzzy C-means Cluster Analysis Research Articles

Wave Height Interval Distribution Method Based on Nonlinear Shallow Water Wave Function

Li, X.; Zhao, Y., and Jin, H., 2020. Wave height interval distribution method based on nonlinear shallow water wave function. In: Yang, Y.; Mi, C.; Zhao, L., and Lam, S. (eds.), Global Topics and New Trends in Coastal Research: Port, Coastal and Ocean Engineering. Journal of Coastal Research, Special Issue No. 103, pp. 843–846. Coconut Creek (Florida), ISSN 0749-0208.In order to improve the ability to analyze and judge the interval distribution of wave height, it is necessary to estimate the interval distribution of wave height. An interval distribution estimation method of wave height based on nonlinear shallow water wave function is proposed. A statistical time series analysis model for estimating the interval distribution of sea wave height is constructed, large data mining and feature extraction for estimating the interval distribution of sea wave height are carried out by adopting a large data feature detection method, ordered clustering of statistical feature sequences of the interval distribution of sea wave height is carried out based on a nonlinear shallow water wave function estimation idea, information clustering and attribute merging in the process of estimating the interval distribution of sea wave height are carried out by combining a fuzzy C-means clustering analysis method. The feature quantity of association rules of statistical time series of wave height interval distribution is extracted, and the association rule set of statistical time series of wave height interval distribution is analyzed by using principal component analysis method, so as to realize accurate estimation of wave height interval distribution quantity in weighted Markov chain. The simulation results show that the method has high accuracy in estimating the interval distribution of wave height and improves the quantitative analysis capability of the interval distribution of wave height.

Bayesian neural network–based thermal error modeling of feed drive system of CNC machine tool

It is well known that thermal error has a significant impact on the accuracy of CNC machine tools. In order to decrease the thermally induced positioning error of machine tools, a novel thermal error modeling approach based on Bayesian neural network is proposed in this paper. The relationship between the temperature rise and positioning error of the feed drive system is investigated by simultaneously measuring the thermal characteristics that include the temperature field and positioning error of the CNC machine tool. Fuzzy c-means (FCM) clustering and correlation analysis are used to select temperature-sensitive points, and the Dunn index is introduced to determine the optimal number of clustering groups, which can inhibit the multicollinearity problem among temperature measuring points effectively. The least-square linear fitting is applied to explore the feature of the positioning error data. The results show that compared with the BP neural network and multiple linear regression model, the Bayesian neural network not only has higher prediction accuracy but also can guarantee excellent prediction performance under different working conditions. The prediction results obtained under different operating conditions indicate that the maximum thermal error can be reduced from around 18.2 to 5.14 μm by using the Bayesian neural network, which represents a 71% reduction in the thermally induced error of the feed drive system of machine tool.

Study on fiber fracture sequence during yarn tensile fracture via acoustic emission method

via the acoustic emission technology, a method used for investigating the single fiber fracture sequence during tensile fracture process of blended yarn, which is made of different kinds of fiber was presented in this paper. As the fibers aggregate together after spinning process, the tensile strength of the yarn mainly originates from the strength of the fibers inside the yarn, as well as the cohesive force and surface friction between the fibers. In order to collect the acoustic emission signals generated during tensile fracture of the yarn and different single fibers, the acoustic emission technology was adopted. Firstly, the time-frequency matrix of the acoustic emission signal was constructed on the basis of Hilbert-Huang transform; subsequently, the singular value of the signal was obtained by singular value decomposition of the time-frequency matrix of the signal; ultimately, the singular value of the signal was judged by the fuzzy c-means cluster analysis. The results showed that the singular value can be used not only to characterize the fractures of different kinds of fiber in an efficient way, but also to identify the fiber fracture sequence during tensile fracture process of blended yarn, providing an excellent theoretical support for researching into the yarn formation process.

Fuzzy Weighted Clustering Method for Numerical Attributes of Communication Big Data Based on Cloud Computing

It is necessary to optimize clustering processing of communication big data numerical attribute feature information in order to improve the ability of numerical attribute mining of communication big data, and thus a big data clustering algorithm based on cloud computing was proposed. The cloud extended distributed feature fitting method was used to process the numerical attribute linear programming of communication big data, and the mutual information feature quantity of communication big data numerical attribute was extracted. Combined with fuzzy C-means clustering and linear regression analysis, the statistical analysis of big data numerical attribute feature information was carried out, and the associated attribute sample set of communication big data numerical attribute cloud grid distribution was constructed. Cloud computing and adaptive quantitative recurrent classifiers were used for data classification, and block template matching and multi-sensor information fusion were combined to search the clustering center automatically to improve the convergence of clustering. The simulation results show that, after the application of this method, the information fusion performance of the clustering process was better, the automatic searching ability of the data clustering center was stronger, the frequency domain equalization control effect was good, the bit error rate was low, the energy consumption was small, and the ability of fuzzy weighted clustering retrieval of numerical attributes of communication big data was effectively improved.

USE OF THE FUZZY CLUSTERING ALGORITHM FOR PATTERN RECOGNITION IN FEED CONSUMPTION DATA OF PURE NEW ZEALAND WHITE RABBITS EXPOSED TO VARIED THERMAL CHALLENGES

The performance of New Zealand White rabbits (NZW) is directly associated with to ambiance-related factors because they present high sensitivity to high-temperature conditions. The objective of the present work was to use the Fuzzy C-Means (FCM) clustering algorithm for pattern recognition in daily feed consumption (CDR) of NZW rabbits exposed to different thermal challenges. The experiment was carried out in four air-conditioned wind tunnels installed in a laboratory. Twenty-four pure rabbits of the NZW breed aged 30 to 37 days were used. The experiment was carried out in two stages with a period of seven days each, and, at each stage, four dry bulb temperatures (20°C, 24ºC, 28ºC and 32ºC) were tested from the 30th day of the rabbits’ life. Data on CDR (kilo, kg day-1) were obtained by weighing the quantities supplied and the leftovers obtained daily from each rabbit in each treatment. Afterward, the Fuzzy C-Means algorithm (FCM) was used to classify the results. Also, to validate the analysis, the validation indexes were applied to indicate in which quantities of clusters the best partition results were obtained for this database. Thus, FCM cluster analysis was set up as a methodology capable of providing information on the thermal comfort of NZB rabbits in a precise and non-invasive way, which could assist the producer in decision-making.

Local segmentation of images using an improved fuzzy C-means clustering algorithm based on self-adaptive dictionary learning

Image segmentation is an active research topic in image processing. The Fuzzy C-means (FCM) clustering analysis has been widely used in image segmentation. As there is a large amount of delicate tissues such as blood vessels and nerves in medical images, noise generated during imaging process can easily affect successful segmentation of these tissues. The traditional FCM algorithm is not ideal for segmentation of images containing strong noise. In this study, we proposed an improved FCM algorithm with anti-noise capability. We first discussed the algorithm of dictionary learning for noise reduction. Then we developed a new image segmentation algorithm as a combination of the dictionary learning for noise reduction and the improved fuzzy C-means clustering. Lastly we used the algorithm of the improved FCM to segment images, during which we removed the non-target areas making use of the grayscale features of images and extracted accurately the areas of interests. The algorithm was tested using synthetic Shepp-Logan images and real medical magnetic resonance imaging (MRI) and computed tomography (CT) images. Compared to the synthetic data and real medical images segmented by the fuzzy C-means (FCM) clustering algorithm, the Kernel Fuzzy C-mean (KFCM) clustering algorithm, spectral clustering algorithm, the sparse learning based fuzzy C-means (SL_FCM) clustering algorithm, and the modified spatial KFCM (MSFCM) algorithm, the images segmented by the dictionary learning Fuzzy C-mean clustering (DLFCM) algorithm have higher partition coefficient, lower partition entropy, better visual perception, better clustering accuracy, and clustering purity.

Unsupervised Clustering Analysis Based on MODS Severity Identifies Four Distinct Organ Dysfunction Patterns in Severely Injured Blunt Trauma Patients.

Purpose: We sought to identify a MODS score parameter that highly correlates with adverse outcomes and then use this parameter to test the hypothesis that multiple severity-based MODS clusters could be identified after blunt trauma.Methods: MOD score across days (D) 2–5 was subjected to Fuzzy C-means Clustering Analysis (FCM) followed by eight Clustering Validity Indices (CVI) to derive organ dysfunction patterns among 376 blunt trauma patients admitted to the intensive care unit (ICU) who survived to discharge. Thirty-one inflammation biomarkers were assayed (Luminex™) in serial blood samples (3 samples within the first 24 h and then daily up to D 5) and were analyzed using Two-Way ANOVA and Dynamic Network analysis (DyNA).Results: The FCM followed by CVI suggested four distinct clusters based on MOD score magnitude between D2 and D5. Distinct patterns of organ dysfunction emerged in each of the four clusters and exhibited statistically significant differences with regards to in-hospital outcomes. Interleukin (IL)-6, MCP-1, IL-10, IL-8, IP-10, sST2, and MIG were elevated differentially over time across the four clusters. DyNA identified remarkable differences in inflammatory network interconnectivity.Conclusion: These results suggest the existence of four distinct organ failure patterns based on MOD score magnitude in blunt trauma patients admitted to the ICU who survive to discharge.

RESEARCH ON EVALUATION METHOD OF AIRBORNE LIDAR POSITIONING ACCURACY

Abstract. Due to the semi-random characteristics of ground points collected by airborne LIDAR system, it is difficult to control the laser pin points to fall on the control points with known coordinates in actual measurement, so the accuracy can not be evaluated by directly comparing coordinate data. In this paper, based on the target plate designed by air-to-ground, the fuzzy c-means clustering analysis algorithm is proposed to extract the point cloud data on the target according to the different echo intensity data of laser on different ground objects. The center point coordinates of the target were fitted by the point cloud data on the target using the method of circumscribed circle of edge points, so as to realize the plane precision and elevation accuracy of airborne LIDAR system of evaluation. The results show that the model fitting method can quickly and effectively evaluate the accuracy of airborne LIDAR, and the method is simple and feasible.

Quantitative Proteomics Reveals the Protective Effects of Huangqi Decoction Against Acute Cholestatic Liver Injury by Inhibiting the NF-κB/IL-6/STAT3 Signaling Pathway.

Intrahepatic cholestasis (IC) is a common syndrome that affects the liver, with treatment options being limited. Huangqi decoction (HQD), a classic herbal medicine, has shown protective effects against IC. In this study, isobaric tags for relative and absolute quantification-based quantitative proteomics was performed to investigate the potential mechanism of action of HQD on α-naphthylisothiocyanate (ANIT)-induced IC, resulting in 2796 quantified proteins across all samples, including 270 differentially expressed proteins under HQD treatment. Fuzzy c-means clustering analysis of these 270 proteins assigned the proinflammatory proteins, such as LCN2, SAA1, FGG, FGA, and FGB, to Cluster 1 (upregulated by ANIT, and downregulated by HQD). Functional bioinformatics and protein-protein interaction network analyses indicated that these proinflammatory proteins were involved in the STAT3 signaling pathway. Further real-time PCR and Western blot experiments confirmed that the expression of these proteins was consistent with the proteomic results. Moreover, HQD treatment decreased the phosphorylation of STAT3, induced by ANIT. Western blot experiments revealed that HQD treatment decreased phosphorylation of NF-κB and downregulated the expression of the inflammatory gene IL-6 and therefore inhibited the IL-6/STAT3 signaling pathway. In summary, the present study suggested that HQD may ameliorate acute cholestatic liver injury via inhibition of the NF-κB/IL-6/STAT3 signaling pathway.

Soft clustering using real-world data for the identification of multimorbidity patterns in an elderly population: cross-sectional study in a Mediterranean population

ObjectivesThe aim of this study was to identify, with soft clustering methods, multimorbidity patterns in the electronic health records of a population ≥65 years, and to analyse such patterns in...

Fuzzy Clustering of Maize Plant-Height Patterns Using Time Series of UAV Remote-Sensing Images and Variety Traits.

The application of high-throughput phenotyping (HTP) techniques based on unmanned aerial vehicle (UAV) remote-sensing platforms to study large-scale population breeding opens the way to more efficient acquisition of dynamic phenotypic traits and provides new tools that should help close the gap between genotyping and traditional field-phenotyping methods. Toward this end we used a field UAV-HTP platform to deploy a RGB high-resolution camera to acquire time-series images. By using three-dimensional reconstructed point cloud models, we developed a repeatable processing workflow to extract plant height from time-series images. The plant height determined by the UAV-HTP platform correlated strongly with that measured manually. The plant heights estimated at various growth stages form temporal profiles that give insights into changes and trends in genotyping. Based on fuzzy c-means clustering analysis, we extract the typical dynamic patterns in phenotypic traits (i.e., plant height, average rate of growth of plant height, and rate of contribution of plant height) hidden in the temporal profiles. The fuzzy c-means clustering and set-intersection operation were first applied to analyze the temporal profile to identify how plant-height patterns change and to detect differences in phenotypic variability among the genotypes. The results revealed the capacity of UAV remote sensing to easily evaluate field traits on multiple timescales, for a few breeding plots or for 1000s of breeding plots.

Magnetic mineral tracing of sediment provenance in the central Bengal Fan

We present a comprehensive magnetic mineral analysis of surface sediments from 110 sites in the central Bengal Fan to trace sediment provenance. Multiple magnetic parameters, including magnetic mineralogy, concentration, and grain size were generated. Rock magnetic measurements indicate that the dominant magnetic minerals within the studied sediments are detrital magnetite and titanomagnetite, which is further confirmed directly by scanning electron microscope (SEM) and transmission electron microscope (TEM) observations on magnetic extracts. Fuzzy c-means clustering analysis of magnetic parameters indicates that the study area can be divided into three clusters. Considering the position of ‘active valley’, we further divide cluster 1 into clusters 1A and 1B. Cluster 1A primarily covers the west side of the ‘active valley’ with relatively high magnetic concentrations and dominantly coarse-grained magnetic minerals, which likely correspond to basaltic materials derived from the Deccan Plateau. From the Godavari river mouth to the inner Bay of Bengal, magnetic concentration gradually decreases with a fining trend of magnetic grain size. Clusters 1B, 2, and 3 are on the east side of the ‘active valley’, where magnetic concentration and magnetic grain size decrease with increasing distance to the ‘active valley’. Combined with published data in the Indian region and the upper Bengal Fan, we suggest that the main provenances are the Himalayan and the Deccan Plateau, whose contributions to sediment distribution are quantitatively assessed. Our analysis suggests that the position of ‘active valley’ plays a key role in sediment distribution in the central Bengal Fan, while the effect of ocean currents is negligible.

Equivalent Modeling of Photovoltaic Power Station Based on Canopy-FCM Clustering Algorithm

Grid-connected photovoltaic power generation has become an effective way to utilize solar energy. In order to accurately analyze the dynamic characteristics of a grid-connected photovoltaic power station, an equivalent modeling method based on the Canopy-FCM clustering algorithm is proposed. Considering the lack of theoretical basis for the selection of the traditional clustering index, this paper deduces the transfer function of the grid-connected inverter control system based on the detailed model and proposes a new clustering index, which takes the zero-pole expression of the closed-loop transfer function as the prototype. In order to improve the fuzzy C-means clustering method, which is sensitive to the initial clustering centers and the outliers, a Canopy algorithm with lower computation cost is introduced. Before fuzzy C-means clustering analysis, the Canopy algorithm is used to determine the initial clustering centers and the number of clusters to improve the accuracy and efficiency of the clustering algorithm. The effectiveness of the proposed method is verified by simulation examples.

An adaptive artificial-fish-swarm-inspired fuzzy C-means algorithm

Fuzzy C-means (FCM) is a classical algorithm of cluster analysis which has been applied to many fields including artificial intelligence, pattern recognition, data aggregation and their applications in software engineering, image processing, IoT, etc. However, it is sensitive to the initial value selection and prone to get local extremum. The classification effect is also unsatisfactory which limits its applications severely. Therefore, this paper introduces the artificial-fish-swarm algorithm (AFSA) which has strong global search ability and adds an adaptive mechanism to make it adaptively adjust the scope of visual value, improves its local and global optimization ability, and reduces the number of algorithm iterations. Then it is applied to the improved FCM which is based on the Mahalanobis distance, named as adaptive AFSA-inspired FCM(AAFSA-FCM). The optimal solution obtained by adaptive AFSA (AAFSA) is used for FCM cluster analysis to solve the problems mentioned above and improve clustering performance. Experiments show that the proposed algorithm has better clustering effect and classification performance with lower computing cost which can be better to apply to every relevant area, such as IoT, network analysis, and abnormal detection.

Delineation of homogeneous regions for streamflow via fuzzy c-means in the Amazon

Abstract Lack of streamflow data is one of the main limitations in hydrologic studies. One method of solving this problem is by streamflow regionalization. The identification of hydrologically homogeneous regions is the main and most important stage of regionalization. In this study homogeneous flow regions are identified by fuzzy c-means (FCM) cluster analysis based on morpho-climatic characteristics from streamflow at 208 stream gauges in the Amazon region. The optimal number of clusters in the dataset was identified by applying the PBM validation index, maximized for ten clusters, with a fuzzing parameter of 1.6. The application dataset is best divided into 10 groups. These were well defined and demonstrated the Amazon's hydrologic similarity.

Analiza oštećenja kompozitnih laminata pri udaru sa niskim brzinama korišćenjem efekta akustične emisije

Characterization of dynamic induced damage is one of the most controversial issues in the application of composite structures. To this aim, Acoustic Emission (AE) technique has been qualified as a robust tool for damage sequence analysis due to its high sensitivity to damage mechanisms. First, AE reference map was created using a couple analysis of Fuzzy C-Means (FCM) clustering and Principal Component Analysis (PCA). The low velocity impact tests were then conducted on composite plates. Finally, Impact AE signals were discriminated and the sequence of damage mechanisms during impact process was discovered. As a result, it is concluded that AE method accompanied with the presented method is a new appropriate approach to discriminate damage mechanism sequences in impacted composite specimens.

Quantitative Proteomic Study Reveals Up-Regulation of cAMP Signaling Pathway-Related Proteins in Mild Traumatic Brain Injury.

Traumatic brain injury (TBI), as a neurological injury, becomes a leading cause of disability and mortality due to lacking effective therapy. About 75% of TBI is mild traumatic brain injury (mTBI). However, the complex molecular mechanisms underlying mTBI pathophysiology remains to be elucidated. In this study, iTRAQ-based quantitative proteomic approach was employed to measure temporal-global proteome changes of rat brain tissues from different time points (1 day, 7 day and 6 months) post single mTBI (smTBI) and repetitive mTBI (rmTBI). A total of 5169 proteins were identified, of which, 237 proteins were significantly changed between control rats and mTBI model rats. Fuzzy c-means (FCM) clustering analysis classified these 237 proteins into six clusters according to their temporal pattern of protein abundance. Functional bioinformatics analysis and protein-protein interaction (PPI) network mapping of these FCM clusters showed that phosphodiesterase 10A (Pde10a) and guanine nucleotide-binding protein G (olf) subunit alpha (Gnal) were the node proteins in the cAMP signaling pathway. Other biological processes, such as cell adhesion, autophagy, myelination, microtubule depolymerization and brain development, were also over-represented in FCM clusters. Further Western Blot experiments confirmed that Pde10a and Gnal were acutely up-regulated in severity-dependent manner by mTBI, but these two proteins could not be down-regulated to basal level at the time point of 6 months post repetitive mTBI. Our study demonstrated that different severity of mTBI cause significant temporal profiling change at the proteomic level and pointed out the cAMP signaling pathway-related proteins, Pde10a and Gnal, may play important roles in the pathogenesis and recovery of mTBI.

Mapping of rock types using a joint approach by combining the multivariate statistics, self-organizing map and Bayesian neural networks: an example from IODP 323 site

Modeling and classification of the subsurface lithology is very important to understand the evolution of the earth system. However, precise classification and mapping of lithology using a single framework are difficult due to the complexity and the nonlinearity of the problem driven by limited core sample information. Here, we implement a joint approach by combining the unsupervised and the supervised methods in a single framework for better classification and mapping of rock types. In the unsupervised method, we use the principal component analysis (PCA), K-means cluster analysis (K-means), dendrogram analysis, Fuzzy C-means (FCM) cluster analysis and self-organizing map (SOM). In the supervised method, we use the Bayesian neural networks (BNN) optimized by the Hybrid Monte Carlo (HMC) (BNN-HMC) and the scaled conjugate gradient (SCG) (BNN-SCG) techniques. We use P-wave velocity, density, neutron porosity, resistivity and gamma ray logs of the well U1343E of the Integrated Ocean Drilling Program (IODP) Expedition 323 in the Bering Sea slope region. While the SOM algorithm allows us to visualize the clustering results in spatial domain, the combined classification schemes (supervised and unsupervised) uncover the different patterns of lithology such of as clayey-silt, diatom-silt and silty-clay from an un-cored section of the drilled hole. In addition, the BNN approach is capable of estimating uncertainty in the predictive modeling of three types of rocks over the entire lithology section at site U1343. Alternate succession of clayey-silt, diatom-silt and silty-clay may be representative of crustal inhomogeneity in general and thus could be a basis for detail study related to the productivity of methane gas in the oceans worldwide. Moreover, at the 530 m depth down below seafloor (DSF), the transition from Pliocene to Pleistocene could be linked to lithological alternation between the clayey-silt and the diatom-silt. The present results could provide the basis for the detailed study to get deeper insight into the Bering Sea’ sediment deposition and sequence.

Coherent tropical-subtropical Holocene see-saw moisture patterns in the Eastern Hemisphere monsoon systems

The concept of a Global Monsoon (GM) has been proposed based on modern precipitation observations, but its application over a wide range of temporal scales is still under debate. Here, we present a synthesis of 268 continental paleo-moisture records collected from monsoonal systems in the Eastern Hemisphere, including the East Asian Monsoon (EAsM), the Indian Monsoon (IM), the East African Monsoon (EAfM), and the Australian Monsoon (AuM) covering the last 18,000 years. The overall pattern of late Glacial to Holocene moisture change is consistent with those inferred from ice cores and marine records. With respect to the last 10,000 years (10 ka), i.e. a period that has high spatial coverage, a Fuzzy c-Means clustering analysis of the moisture index records together with “Xie-Beni” index reveals four clusters of our data set. The paleoclimatic meaning of each cluster is interpreted considering the temporal evolution and spatial distribution patterns. The major trend in the tropical AuM, EAfM, and IM regions is a gradual decrease in moisture conditions since the early Holocene. Moisture changes in the EAsM regions show maximum index values between 8 and 6 ka. However, records located in nearby subtropical areas, i.e. in regions not influenced by the intertropical convergence zone, show an opposite trend compared to the tropical monsoon regions (AuM, EAfM and IM), i.e. a gradual increase. Analyses of modern meteorological data reveal the same spatial patterns as in the paleoclimate records such that, in times of overall monsoon strengthening, lower precipitation rates are observed in the nearby subtropical areas. We explain this pattern as the effect of a strong monsoon circulation suppressing air uplift in nearby subtropical areas, and hence hindering precipitation. By analogy to the modern system, this would mean that during the early Holocene strong monsoon period, the intensified ascending airflows within the monsoon domains led to relatively weaker ascending or even descending airflows in the adjacent subtropical regions, resulting in a precipitation deficit compared to the late Holocene. Our conceptual model therefore integrates regionally contrasting moisture changes into the Global Monsoon hypothesis.

On urban road traffic state evaluation index system and method

Traffic state evaluation is a basic and critical work in the research on road traffic congestion. It can provide basic data support for the improvement measures and information release in traffic management and service. The aim of this research is to obtain a comprehensive value to describe traffic state accurately based on the evaluation index system. In this paper, it is carried out using fuzzy c-means (FCM) algorithm and fuzzy entropy weight method. In the framework, traffic flow was classified into six different states to determine the fuzzy range of indices using the improved FCM clustering analysis. Besides, fuzzy entropy weight method is proposed to compute the evaluation result of traffic state for section, road and road network, respectively. The experiments based on the traffic information in a subset of Beijing’s road network prove that the findings of traffic evaluation are in accordance with the actual situation and people’s sense of traffic state.