Fuzzy Relational Clustering Research Articles

Cophenetic-based fuzzy clustering of time series by linear dependency

In this work, a new approach to cluster large sets of time series is presented. The proposed methodology takes into account the dependency among the time series to obtain a fuzzy partition of the set of observations. A two-step procedure to accomplish this is presented. First, the cophenetic distances, based on a time series linear cross-dependency measure, are obtained. Second, these distances are used as an input of a non-Euclidean fuzzy relational clustering algorithm. As a result, we obtain a robust fuzzy procedure capable of detecting groups of time series with different types of cross-dependency. We illustrate the usefulness of the stated methodology through some Monte Carlo experiments and a real data example. Our results show that the methodology proposed in this work substantially improves the hard partitioning clustering alternative.

Automatic Determination of the Number of Clusters for Semi-Supervised Relational Fuzzy Clustering

Semi-supervised clustering relies on both labeled and unlabeled data to steer the clustering process towards optimal categorization and escape from local minima. In this paper, we propose a novel fuzzy relational semi-supervised clustering algorithm based on an adaptive local distance measure (SSRF-CA). The proposed clustering algorithm utilizes side-information and formulates it as a set of constraints to supervise the learning task. These constraints are expressed using reward and penalty terms, which are integrated into a novel objective function. In particular, we formulate the clustering task as an optimization problem through the minimization of the proposed objective function. Solving this optimization problem provides the optimal values of different objective function parameters and yields the proposed semi-supervised clustering algorithm. Along with its ability to perform data clustering and learn the underlying dissimilarity measure between the data instances, our algorithm determines the optimal number of clusters in an unsupervised manner. Moreover, the proposed SSRF-CA is designed to handle relational data. This makes it appropriate for applications where only pairwise similarity (or dissimilarity) information between data instances is available. In this paper, we proved the ability of the proposed algorithm to learn the appropriate local distance measures and the optimal number of clusters while partitioning the data using various synthetic and real-world benchmark datasets that contain varying numbers of clusters with diverse shapes. The experimental results revealed that the proposed SSRF-CA accomplished the best performance among other state-of-the-art algorithms and confirmed the outperformance of our clustering approach.

PSO for Fuzzy Clustering of Multi-view Relational Data

Particle Swarm Optimization (PSO) is a population-based meta-heuristic known for its simplicity, being successfully used in clustering task with interesting performance. Clustering of multi-view data sets has received increasing attention since it explores multiple sources or views of data sets aiming at improving clustering accuracy. Previous studies mainly focused on PSO-based clustering of single-view vector data, neither single- nor multi-view PSO-based clustering of relational received proper attention. This paper introduces a PSO-based approach to the fuzzy clustering of multi-view relational data, which can cluster data sets described by several dissimilarity matrices, each of them representing a particular view. In this work, ten fitness functions were considered, in which eight of them were adapted to deal with multi-view relational data and to consider the relevance weights of views. These fitness functions were compared to evaluate which best fit to cluster multi-view relational data. The performance and usefulness of the proposed approach, in comparison with previous single- and multi-view relational fuzzy clustering algorithms, are illustrated with several multi-view data sets. The Adjusted Rand Index (ARI) and F-measure were used to assess the quality of fuzzy partitions provided by clustering algorithms. The results have shown that the proposed methods significantly outperformed the compared algorithms in the majority of cases.

A Real-Time Self-Healing Methodology Using Model- and Measurement-Based Islanding Algorithms

In this paper, a new real-time defensive islanding method, which is adaptive to the operating conditions is proposed. In the method, a number of candidate islanding schemes are generated using both model- and measurement-based islanding algorithms after detecting a severe fault in the system by means of a new severity index based on generator bus voltage frequency measurements. Of model-based algorithms, slow coherency-based islanding, in which the prefault measurements are utilized, is adopted. On the other hand, ${K}$ -means, hierarchical, and fuzzy relational eigenvector centrality-based clustering are employed as measurement-based islanding algorithms, where the postfault measurements of the evolving dynamics after the severe fault are utilized. A faster-than-real-time software platform is, then, employed to validate the success of the candidate schemes in healing the system. Among the successful schemes, the one resulting in the least load imbalance is chosen to be applied. All the computations from the detection of the fault to the application of islanding are performed in real-time, directly after the occurrence of the fault. The proposed method is demonstrated on the 37-generator 127-bus WSCC power test system, and on a model of the Turkish power system to assess the method’s performance.

Improving the Efficiency and the Accuracy of 2D Gel Electrophoresis Spot Detection Using Graphics Processing Units

Background: Proteomics was built around two-dimensional (2D) gel electrophoresis. Accurately analyzing the images generated from 2D gel electrophoresis for spot detection is a timeconsuming process especially for high-resolution and large images. Objective: In this paper, we present an accurate GPU-accelerated software tool for the detection and quantification of protein spots in 2D gel electrophoresis images. Method: We adopt pixel-based approach that employs wavelet relational fuzzy C-means clustering and distance transform to detect and quantify the protein spots. This pixel-based spot detection approach is more accurate than the contour-based approaches; however it is compute-intensive. So, along with algorithmic optimizations, we present the mapping and optimization of the pixel-based spot detection algorithm onto graphics processing units (GPUs); including NVIDIA and AMD GPUs. Results: This approach is proved to exhibit better spot detection in quantitative comparisons with the commercial software tools. Specifically, it achieves a degree of improvement in F-measure of 21.237% and 11.716% on the average compared to Delta2D and Melanie, respectively. We carry out experiments on images of large size and high resolution for healthy and diseased samples. Our implementation has accomplished up to five orders of magnitude speedup compared to the single-threaded MATLAB implementation. Conclusion: We proposed an accurate and efficient tool for detecting and quantifying the protein spots in 2D Gel Electrophoresis images. Our tool outperforms commercial software tools in accuracy of detecting protein spots while achieving significantly better performance than single-threaded MATLAB implementation by utilizing the GPU accelerators. Keywords: Image processing, 2D gel electrophoresis, heterogeneous computing, parallel computing, GPU, OpenCL.

New automatic fuzzy relational clustering algorithms using multi-objective NSGA-II

Most of the fuzzy clustering methods use only centroid information and cannot differentiate the geometric structures of clusters due to the cohesion and separation measures of the fuzzy partition. Moreover, conventional clustering methods confine the search space of automatic fuzzy clustering as it shows a tendency to fall into local minima as well as the number of clusters is required to be provided as prior knowledge. In this paper, combining fuzzy relational clustering (FRC) with multi-objective genetic algorithm NSGA-II, we propose two new fuzzy relational clustering methods, referred to as FRCNSGA and IFRCNSGA. In IFRCNSGA, NSGA-II is used to adjust the parameters of FRC algorithm dynamically such as the appropriate number of clusters and the initial membership values. Thus this proposed model handles the intricacy of automatic clustering a dataset, at the same time, avoids the problem of falling into local minima quickly. For both of the proposed models, NSGA-II is used to optimize two cluster validity indices named, separation, and cohesion, simultaneously, whereas, FRC is used to handle the overlapping properties of clusters. In this case, fuzzy membership degrees are used to compute the overlap-separation. To eliminate the impediment of automatic fuzzy clustering, it is necessary to optimize one or more criteria. Therefore, we consider NSGA-II as an optimization algorithm to optimize more than single criterion. The experimental results have exhibited that the intended methods provide competitive results in the compound, overlapped, high-dimensional gene expression and non-gene expression datasets. The proposed methods are able to determine well-separated, hyperspherical, noncompact and overlapping clusters compared to other existing methods.

Optimal sentence clustering for web database using hierarchical fuzzy relational clustering integrated with artificial bee colony algorithm

Optimal sentence clustering for web database using hierarchical fuzzy relational clustering integrated with artificial bee colony algorithm

Optimal Sentence Clustering for Web Database using Hierarchical Fuzzy Relational Clustering Integrated with Artificial Bee Colony Algorithm

Sentence clustering plays a vital role in text mining and text processing activities. The proposed work is a novel hierarchical fuzzy relational clustering algorithm (HFRECA) capable of identifying sub clusters. It has features of both hierarchical clustering and fuzzy clustering in which it uses page rank to form multiple clusters present in text documents containing hierarchical structure. To enhance the quality of the clusters formed, an optimisation algorithm which is called artificial bee colony (ABC) algorithm is used along with it. The proposed algorithm identifies the sub clusters and finely tunes the cluster to show a better optimised result.

Incorporating gene ontology into fuzzy relational clustering of microarray gene expression data

The product of gene expression works together in the cell for each living organism in order to achieve different biological processes. Many proteins are involved in different roles depending on the environment of the organism for the functioning of the cell. In this paper, we propose gene ontology (GO) annotations based semi-supervised clustering algorithm called GO fuzzy relational clustering (GO-FRC) where one gene is allowed to be assigned to multiple clusters which are the most biologically relevant behavior of genes. In the clustering process, GO-FRC utilizes useful biological knowledge which is available in the form of a gene ontology, as a prior knowledge along with the gene expression data. The prior knowledge helps to improve the coherence of the groups concerning the knowledge field. The proposed GO-FRC has been tested on the two yeast (Saccharomyces cerevisiae) expression profiles datasets (Eisen and Dream5 yeast datasets) and compared with other state-of-the-art clustering algorithms. Experimental results imply that GO-FRC is able to produce more biologically relevant clusters with the use of the small amount of GO annotations.

A subtractive medoids selection based fuzzy relational clustering of augmented web user sessions

A subtractive medoids selection based fuzzy relational clustering of augmented web user sessions

A fuzzy similarity measure based on equivalence relation with application in cluster analysis

We propose a novel notion of fuzzy similarity measure between fuzzy sets by using fuzzy equivalence. Furthermore, a similarity base fuzzy relational clustering algorithm is formulated. Applications are given to illustrate the usefulness of the chosen fuzzy equivalence and similarity measure in a partition algorithm.

Comparing clustering models in bank customers: Based on Fuzzy relational clustering approach

Article history: Received December 5, 2015 Received in revised format February 16 2016 Accepted August 15 2016 Available online August 16 2016 Clustering is absolutely useful information to explore data structures and has been employed in many places. It organizes a set of objects into similar groups called clusters, and the objects within one cluster are both highly similar and dissimilar with the objects in other clusters. The K-mean, C-mean, Fuzzy C-mean and Kernel K-mean algorithms are the most popular clustering algorithms for their easy implementation and fast work, but in some cases we cannot use these algorithms. Regarding this, in this paper, a hybrid model for customer clustering is presented that is applicable in five banks of Fars Province, Shiraz, Iran. In this way, the fuzzy relation among customers is defined by using their features described in linguistic and quantitative variables. As follows, the customers of banks are grouped according to K-mean, C-mean, Fuzzy C-mean and Kernel K-mean algorithms and the proposed Fuzzy Relation Clustering (FRC) algorithm. The aim of this paper is to show how to choose the best clustering algorithms based on density-based clustering and present a new clustering algorithm for both crisp and fuzzy variables. Finally, we apply the proposed approach to five datasets of customer's segmentation in banks. The result of the FCR shows the accuracy and high performance of FRC compared other clustering methods. Growing Science Ltd. All rights reserved. 7 © 201

Crack Detection and Identification Using Vibration Signals and Fuzzy Clustering

Cracks in machine and structural components are among the most critical faults that require immediate attention by operations and maintenance personnel. In many cases it is difficult to directly and accurately identify the fault type and its extent under varying operating conditions. Therefore the ability to monitor and diagnose machine and structures health at early stages of fault initiation is one of the fundamental requirements of the development of online machine monitoring systems. This work demonstrates a novel procedure for crack detection and identification in an experimental set-up. Cracks of different lengths are introduced at three different locations along a cantilevered steel beam. The beam is exposed to several forms of excitation at different levels of amplitude and frequency. The excitation is introduced to the beam next to its fixed end using an electromagnetic shaker controlled by an arbitrary function generator. The vibration signals are analyzed for the beam dynamic signatures in the form of statistical moments, frequency spectra and wavelet coefficients. These are then used as features to create a data-partition using fuzzy relational clustering. The objective is to correlate crack length and crack propagation to the statistical moments, frequency spectra and wavelet coefficient data. The work will be later extended to supervise learning by training a classifier in a two-class problem to identify signal features that correlate to critical faults

Fuzzy C-Means Clustering Based on Improved Marked Watershed Transformation

Currently, the fuzzy c-means algorithm plays a certain role in remote sensing image classification. However, it is easy to fall into local optimal solution, which leads to poor classification. In order to improve the accuracy of classification, this paper, based on the improved marked watershed segmentation, puts forward a fuzzy c-means clustering optimization algorithm. Because the watershed segmentation and fuzzy c-means clustering are sensitive to the noise of the image, this paper uses the adaptive median filtering algorithm to eliminate the noise information. During this process, the classification numbers and initial cluster centers of fuzzy c-means are determined by the result of the fuzzy similar relation clustering. Through a series of comparative simulation experiments, the results show that the method proposed in this paper is more accurate than the ISODATA method, and it is a feasible training method.

Quantitative Evaluation of Web user Session Dissimilarity measures using Medoids based Relational Fuzzy clustering

Background/Objectives: Proficient relational clustering of web users’ sessions not only depends on clustering algorithm’s character but also profoundly influenced by the used dissimilarity measures. Therefore, determining the right dissimilarity measure to capture the actual access behaviour of the web user is imperative for the significant clustering.Methods: In this paper, the concept of an augmented session is used to derive different augmented session dissimilarity measures. The quantitative performance evaluation of different session dissimilarity measures are performed using a relational fuzzy c-medoid clustering approach. The intra-cluster and inter-cluster distance based cluster quality ratio is used for performance evaluation. Findings: The experimental results demonstrated that augmented web user session dissimilarity in general, and intuitive augmented session dissimilarity, in particular, performed better than the other dissimilarity measures. Improvements: It is argued that augmented session similarity measures are more realistic and represent session similarities based on the web user’s habits, interest, and expectations as compared to simple binary session similarity measures.

A Discounted Fuzzy Relational Clustering of Web Users’ Using Intuitive Augmented Sessions Dissimilarity Metric

Relational fuzzy clustering (RFC) algorithms prove very useful in Web user session clustering because Web user sessions may contain fuzzy, conflicting and imprecise information. Though RFC algorithms are very sensitive to cluster initialization and works only if the numbers of clusters are specified in advance. However, at all times, the prior initialization of a number of clusters is not feasible due to the dynamically evolving nature of user sessions. Therefore, estimating the number of clusters and initializing suitable cluster prototype are a significant performance bottleneck in this method. In this paper, the discounted fuzzy relational clustering (DFRC) algorithm is proposed to address the major constraint of RFC. The DFRC algorithm identifies Web user session clusters from Web server access logs, without initializing the number of clusters and prototypes of initial clusters. The DFRC algorithm works in two stages. In the first stage, DFRC automatically identifies the number of potential clusters based on the successively discounted potential density function value of each relational data and their respective centres. In the second stage, DFRC assigns fuzzy membership values to each data point and forms fuzzy clusters from the relational matrix. The DFRC algorithm is applied on an augmented session dissimilarity matrix obtained from a publicly accessed NASA Web server log data. The experimental results are evaluated using different fuzzy validity measures. The extensive experiments are performed to test the effect of various parameters, including accept/reject ratio and neighbourhood radius on the performance of DFRC algorithm. The results were also compared with fuzzy relational clustering algorithm using cluster quality measures. It is observed that the quality of generated clusters using DFRC is superior as compared with that of RFC.

Fuzzy Distance Measure and Fuzzy Clustering Algorithm

A modified dissimilarity measure for fuzzy data is proposed and is applied to real data with mixed feature variables of symbolic and fuzzy data. We also proposed an improved version of fuzzy relational clustering algorithm. Numerical examples and comparison are given between mixed variable fuzzy C-means and fuzzy relational clustering algorithm.

Aggregated Probabilistic Fuzzy Relational Sentence Level Expectation Maximization Clustering Algorithm for Efficient Text Categorization

Now a days, Text clustering becomes an important application to organize the data and to extract useful information from the available corpus. Many previous clustering techniques have difficulties in handling extreme outliers but fuzzy clustering algorithms tend to give them very small membership degree in surrounding clusters. In this paper we proposed an aggregated probabilistic Fuzzy relational sentence level expectation maximization clustering algorithm for efficient text categorization. It will give the accurate and maximum similarity by finding the relevance of sentences which belongs to a particular cluster. This technique leads to a fuzzy partition of the sentences and find out the accurate probability of the words belongs to a cluster. This algorithm is particularly used in finding maximum likelihood estimates of words in a given sentence. It gives the low search results with highest accuracy. The practical results show that the proposed method obtains better and accurate results for getting best sentence-wise text classification when compared with the existing methods.

Fuzzy Co-Clustering Algorithms Based on Fuzzy Relational Clustering and TIBA Imputation

In this paper, two types of fuzzy co-clustering algorithms are proposed. First, it is shown that the base of the objective function for the conventional fuzzy co-clustering method is very similar to the base for entropy-regularized fuzzy nonmetric model. Next, it is shown that the non-sense clustering problem in the conventional fuzzy co-clustering algorithms is identical to that in fuzzy nonmetric model algorithms, in the case that all dissimilarities among rows and columns are zero. Based on this discussion, a method is proposed applying entropy-regularized fuzzy nonmetric model after all dissimilarities among rows and columns are set to some values using a TIBA imputation technique. Furthermore, since relational fuzzy cmeans is similar to fuzzy nonmetricmodel, in the sense that both methods are designed for homogeneous relational data, a method is proposed applying entropyregularized relational fuzzyc-means after imputing all dissimilarities among rows and columns with TIBA. Some numerical examples are presented for the proposed methods.

Fuzzy Relational Clustering Based on Knowledge Mesh and its Application

A selection method of knowledge meshes based on fuzzy relational clustering is proposed. Considering the perfection degree, the matching degree among knowledge meshes and the level frame of knowledge mesh, the similarity function is defined. Its properties are proved. The similarity values between knowledge meshes are regarded as clustering data. The fuzzy relational matrix is constructed and decomposed. The knowledge meshes with high membership in each class are regarded as referenced knowledge meshes. Or the knowledge meshes in the class are further chosen according to users needs. Finally the example shows that the method is effective.