In the distributed detection system with multiple sensors, there are two ways for local sensors to deliver their local decisions to the fusion center (Fc): a one-bit hard decision and a multiple-bit soft decision. com...
详细信息
In the distributed detection system with multiple sensors, there are two ways for local sensors to deliver their local decisions to the fusion center (Fc): a one-bit hard decision and a multiple-bit soft decision. compared with the soft decision, the hard decision has worse detection performance due to the loss of sensing information but has the main advantage of smaller communication costs. To get a tradeoff between communication costs and detection performance, we propose a soft-hard combination decision fusion scheme for the clustered distributed detection system with multiple sensors and non-ideal communication channels. A clustered distributed detection system is configured by a fuzzy logic system and a fuzzyc-meansclustering algorithm. In clusters, each local sensor transmits its local multiple-bit soft decision to its corresponding cluster head (cH) under the non-ideal channel, in which a simple and efficient soft decision fusion method is used. Between clusters, the fusion center combines all cluster heads' one-bit hard decisions into a final global decision by using an optimal fusion rule. We show that the clustered distributed system with the proposed scheme has a good performance that is close to that of the centralized system, but it consumes much less energy than the centralized system at the same time. In addition, the system with the proposed scheme significantly outperforms the conventional distributed detection system that only uses a hard decision fusion. Using simulation results, we also show that the detection performance increases when more bits are delivered in the soft decision in the distributed detection system.
Purpose - Data envelopment analysis (DEA) is a non-parametric model that is developed for evaluating the relative efficiency of a set of homogeneous decision-making units that each unit transforms multiple inputs into...
详细信息
Purpose - Data envelopment analysis (DEA) is a non-parametric model that is developed for evaluating the relative efficiency of a set of homogeneous decision-making units that each unit transforms multiple inputs into multiple outputs. However, usually the decision-making units are not completely similar. The purpose of this paper is to propose an algorithm for DEA applications when considered DMUs are non-homogeneous. Design/methodology/approach - To reach this aim, an algorithm is designed to mitigate the impact of heterogeneity on efficiency evaluation. Using fuzzy c-means algorithm, a fuzzyclustering is obtained for DMUs based on their inputs and outputs. Then, the fuzzyc-means based DEA approach is used for finding the efficiency of DMUs in different clusters. Finally, the different efficiencies of each DMU are aggregated based on the membership values of DMUs in clusters. Findings - Heterogeneity causes some positive impact on some DMUs while it has negative impact on other ones. The proposed method mitigates this undesirable impact and a different distribution of efficiency score is obtained that neglects this unintended impacts. Research limitations/implications - The proposed method can be applied in DEA applications with a large number of DMUs in different situations, where some of them enjoyed the good environmental conditions, while others suffered from bad conditions. Therefore, a better assessment of real performance can be obtained. Originality/value - The paper proposed a hybrid algorithmcombination of fuzzyc-meansclustering method with classic DEA models for the first time.
fuzzyclustering is useful to mine complex and multi-dimensional data sets, where the members have partial or fuzzy relations. Among the various developed techniques, fuzzy-c-means (FcM) algorithm is the most popular ...
详细信息
fuzzyclustering is useful to mine complex and multi-dimensional data sets, where the members have partial or fuzzy relations. Among the various developed techniques, fuzzy-c-means (FcM) algorithm is the most popular one, where a piece of data has partial membership with each of the pre-defined cluster centers. Moreover, in FcM, the cluster centers are virtual, that is, they are chosen at random and thus might be out of the data set. The cluster centers and membership values of the data points with them are updated through some iterations. On the other hand, entropy-based fuzzyclustering (EFc) algorithm works based on a similarity-threshold value. contrary to FcM, in EFc, the cluster centers are real, that is, they are chosen from the data points. In the present paper, the performances of these algorithms have been compared on four data sets, such as IRIS, WINES, OLITOS and psychosis (collected with the help of forty doctors), in terms of the quality of the clusters (that is, discrepancy factor, compactness, distinctness) obtained and their computational time. Moreover, the best set of clusters has been mapped into 2-D for visualization using a self-organizing map (SOM).
In this paper,K-means and fuzzyc-meansclustering algorithms are used as typical algorithms for data mining *** addition,it uses electricity monthly electricity load data for a certain region in domestic as the testi...
详细信息
ISBN:
(纸本)9781510870994
In this paper,K-means and fuzzyc-meansclustering algorithms are used as typical algorithms for data mining *** addition,it uses electricity monthly electricity load data for a certain region in domestic as the testing *** the MATLAB environment,the clustering clusters and different clustering centers for different power consumption behaviors of power consumes are *** analyzing the clustering clusters and clustering centers,the clusters obtained by different algorithms can be roughly divided into three categories,and the power consumers' electricity habits in this area can also be divided into three *** paper uses the approach to improve students' interest in learning,expand students' practical applications in clustering algorithms,and cultivate students' creative creativity.
In GK-algorithm, fuzzyclustering algorithm with preserved volume was used. However, the added fuzzycovariance matrices in their distance measure were not directly derived from the objective function. A fuzzyc-means...
详细信息
In GK-algorithm, fuzzyclustering algorithm with preserved volume was used. However, the added fuzzycovariance matrices in their distance measure were not directly derived from the objective function. A fuzzy c-means algorithm based on Mahalanobis distance(FcM-HM) was proposed to improve those limitations of GG and GK algorithms, but it is not stable enough when some of its covariance matrices are not equal. The singular problem and the selecting initial values problem are improved. We pointed out that the initial memberships of fuzzyc-mean algorithm which was based on Mahalanobis distance algorithm and the traditional fuzzy c-means algorithm(FcM) algorithmcan't be all equal. The other important issue is how to approach the global minimum value that can improve the cluster accuracy. The methods to detect the local extreme value were developed by this paper. Focusing attention to these two faults, an improved new algorithm, "fuzzyc-means based on Particle Swarm Optimization with Mahalanobis distance(PSO-FcM-HM)", is proposed. We have two aims and goals of our research summary. One is to compare the classification accuracies of fuzzyclustering algorithms based on Mahalanobis distances and Euclidean distances. The other is to choose the initial membership to promote the classification accuracies.
In this paper, we report experimental results of hybrid system using Hidden Markov Models/Multi-Layer Perceptron (HMM/MLP) model as acoustic model and based on the fuzzyc-means (FcM) clustering with optimization with...
详细信息
ISBN:
(纸本)9781509064656
In this paper, we report experimental results of hybrid system using Hidden Markov Models/Multi-Layer Perceptron (HMM/MLP) model as acoustic model and based on the fuzzyc-means (FcM) clustering with optimization with Geneticalgorithm (GA). In this context, we use the result of FcM clustering as initial population of GA, this allows training the GA with a population of empirically generated chromosomes and not randomly initialized. Our results on speech recognition tasks show an increase in the estimates of the posterior probabilities of the correct words after training. We demonstrate the effectiveness of the proposed clustering approach in large-vocabulary speaker-independent continuous speech recognition with regard to the three baseline systems : Discrete HMM, hybrid HMM/MLP with K-means and FcM clustering.
The fuzzyc-means method is investigated to cluster the heavy tailed data by using some measures of distance. A comparison study is provided based on time and precision. The results show that when using the Euclidean ...
详细信息
ISBN:
(纸本)9781509040087
The fuzzyc-means method is investigated to cluster the heavy tailed data by using some measures of distance. A comparison study is provided based on time and precision. The results show that when using the Euclidean distance, the time required is less than if we used Manhattan distance, but the precision is higher when using the Manhattan distance.
The status monitoring data of wind turbines have large, multi-source, heterogeneous, complex and rapid growth of large data characteristics. The existing data processing methods are difficult to guarantee efficiency w...
详细信息
ISBN:
(纸本)9781509067596
The status monitoring data of wind turbines have large, multi-source, heterogeneous, complex and rapid growth of large data characteristics. The existing data processing methods are difficult to guarantee efficiency when handling massive amounts of data, and may miss the best time to troubleshoot. How to deal with the monitoring data more efficiently is of great significance to the accurate judgment of the fault. This paper proposes the use of cloud platform to deal with massive data to improve efficiency. Firstly, the state monitoring model of wind turbine is put forward. Then, the fuzzycmeansclustering algorithm is introduced, and the algorithm process is realized by MapReduce model. Finally, the experiment is carried out with Hadoop platform, using distributed database HBase to store data, and using distributed programming framework MapReduce to calculate data. It is found that with the increase of the data volume and the number of nodes, the cloud platform is able to store and calculate data at a faster speed.
Water bodies identification using multispectral images is a very useful application of image processing. This paper proposed a novel method for water bodies identification from multispectral images using Gabor filter,...
详细信息
ISBN:
(纸本)9781509061358
Water bodies identification using multispectral images is a very useful application of image processing. This paper proposed a novel method for water bodies identification from multispectral images using Gabor filter, fuzzyc-means and canny edge detection algorithm. Gabor filter is a combination of lowpass filter and bandpass filter. This two filters extracting the importance features from satellite images. From the extracted features fuzzy c-means algorithmclustered the various land use and land cover classes. Finally water bodies are identified from land use and land cover classes with the use of canny edge detection methods. The proposed approach was experimented with the use of Landsat7, Landsat-8 satellite images. Our experimental results proved that proposed methods provides better result for water identification with high efficiency.
This paper describes an automatic segmentation approach for PET and T1-weighted MR images using a possibilisticclustering algorithm for deriving fuzzy tissue maps of white matter, gray matter and cerebrospinal fluid ...
详细信息
ISBN:
(纸本)9781538618424
This paper describes an automatic segmentation approach for PET and T1-weighted MR images using a possibilisticclustering algorithm for deriving fuzzy tissue maps of white matter, gray matter and cerebrospinal fluid volumes, and using the fuzzy c-means algorithm for the centers initialization process;this hybrid technique allows to compute the degree of membership of each voxel to different brain tissues. The fuzzy process is illustrated for Alzheimer's disease using phantom images from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Our method, inspired from the conventional possibilisticalgorithm, is less sensitive to noise while taking into consideration the effect of partial volume.
暂无评论