Search

Automatic clustering and dimension reduction are two of the most intriguing topics in the field of clustering. Affinity propagation (AP) is a representative graph-based clustering algorithm in unsupervised learning. However, extracting features from high-dimensional data and providing satisfactory clustering results is a serious challenge for the AP algorithm. Besides, the clustering performance of the AP algorithm is sensitive to preference. In this paper, an improved affinity propagation based on optimization of preference (APBOP) is proposed for automatic clustering on high-dimensional data. This method is optimized to solve the difficult problem of determining the preference of affinity propagation and the poor clustering effect for non-convex data distribution. First, t-distributed stochastic neighbor embedding is introduced to reduce the dimensionality of the original data to solve the redundancy problem caused by excessively high dimensionality. Second, an improved hybrid equilibrium optimizer based on the crisscross strategy (HEOC) is proposed to optimize preference selection. HEOC introduces the crisscross strategy to enhance local search and convergence efficiency. The benchmark function experiments indicate that the HEOC algorithm has better accuracy and convergence rate than other swarm intelligence algorithms. Simulation experiments on high-dimensional and real-world datasets show that APBOP has better effectiveness. [ABSTRACT FROM AUTHOR]

In spite of the impressive diversity of models of fuzzy forecasting, there is still a burning need to arrive at models that are both accurate and highly interpretable. This study proposes a new fuzzy forecasting model designed with the use of the two key techniques, namely clustering and axiomatic fuzzy set (AFS) classification. First, clustering algorithm is utilized to generate clustering-based intervals. Second, the fuzzy trend labeled training data set is constructed based on fuzzy logic relationships and fuzzy trends of historical samples. Then, the AFS classification is exploited to yield the semantic interpretation of each fuzzy trend. The main novelty is that the proposed model not only predicts the value but can also capture the trend prevailing in the time series, and obtain its semantic interpretation. The Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX), inventory demand, and Spanish electricity prices are used in a series of experiments. The results show that the proposed model has both good interpretability and accuracy. [ABSTRACT FROM AUTHOR]

Cluster analysis is important in data mining, especially if there is unsupervised data. Recently, many clustering methods have been proposed. Unfortunately, most of these require the definition of the number of clusters, in advance. This study addresses this weakness by proposing a new automatic clustering algorithm: automatic kernel clustering with bee colony optimization (AKC-BCO). AKC-BCO determines the appropriate number of clusters and assigns data points to correct clusters. This is accomplished by the kernel function, which increases clustering capability. This method is validated using several benchmark data sets. The result is compared with several existing automatic clustering methods. The experiment results demonstrate that the proposed AKC-BCO is more stable and accurate than others. Furthermore, the proposed method is also applied to a real-world medical problem. [ABSTRACT FROM AUTHOR]

Abstract: Inspired by the principle of gene transposon proposed by Barbara McClintock, a new immune computing algorithm for automatic clustering named as Gene Transposon based Clone Selection Algorithm (GTCSA) is proposed in this paper. The proposed algorithm does not require a prior knowledge of the number of clusters; an improved variant of the clonal selection algorithm is used to determine the satisfied number of clusters and the appropriate partitioning of the data set as well. In addition, a novel operation called antibody gene transposon is introduced to the framework of clonal selection algorithm which can realize to find the satisfied number of cluster automatically. The proposed method has been extensively compared with iterated local search approach (ILS) and three well-known automatic clustering algorithms, including automatic clustering using an improved differential evolution algorithm (ACDE); variable-string-length genetic algorithm based clustering techniques (VGA) and the dynamic clustering approach based on particle swarm optimization (DCPSO). 23 datasets with widely varying characteristics are used to demonstrate the superiority of the GTCSA. In addition, GTCSA is applied to a real world application, namely natural image segmentation, with a good performance obtained. [Copyright &y& Elsevier]

This paper introduces niching particle swarm optimization (nichePSO) into clustering analysis and puts forward a clustering algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and increases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clustering problem, which can be useful for the niches' forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters automatically and accurately. [ABSTRACT FROM AUTHOR]

Abstract: A novel non-parametric clustering method based on non-parametric local shrinking is proposed. Each data point is transformed in such a way that it moves a specific distance toward a cluster center. The direction and the associated size of each movement are determined by the median of its K-nearest neighbors. This process is repeated until a pre-defined convergence criterion is satisfied. The optimal value of the number of neighbors is determined by optimizing some commonly used index functions that measure the strengths of clusters generated by the algorithm. The number of clusters and the final partition are determined automatically without any input parameter except the stopping rule for convergence. Experiments on simulated and real data sets suggest that the proposed algorithm achieves relatively high accuracies when compared with classical clustering algorithms. [Copyright &y& Elsevier]