Possibilistic Exponential Fuzzy Clustering

IEEE Trans. Fuzzy Syst.

2016

In this paper a novel possibilistic c-means clustering algorithm, called Adaptive Possibilistic c-means, is presented. Its main feature is that its parameters, after their initialization, are properly adapted during its execution. Provided that the algorithm starts with a reasonable overestimate of the number of physical clusters formed by the data, it is capable, in principle, to unravel them (a long-standing issue in the clustering literature). This is due to the fully adaptive nature of the proposed algorithm that enables the removal of the clusters that gradually become obsolete. In addition, the adaptation of all its parameters increases the flexibility of the algorithm in following the variations in the formation of the clusters that occur from iteration to iteration. Theoretical results that are indicative of the convergence behavior of the algorithm are also provided. Finally, extensive simulation results on both synthetic and real data highlight the effectiveness of the proposed algorithm.

Section: B Pcm Issues and Potential Solutionsmentioning

confidence: 99%

“…In [21] ideas from [12] and [11] are combined for dealing additionaly with the coincident clusters issue. The same issues are also addressed in [13] using, however, a different approach than [21].…”

Section: Introductionmentioning

confidence: 99%

A Novel Adaptive Possibilistic Clustering Algorithm

IEEE Trans. Fuzzy Syst.

2016

“…Specifically, in order to compare a clustering outcome with the true data label information, we use (a) the Rand Measure (RM) (e.g. [1]), which measures the degree of agreement between the obtained clustering and the physical clustering and can handle clusterings whose number of clusters may differ from the number of physical clusters, (b) the Success Rate (SR), which measures the percentage of the points that have been correctly labeled by an algorithm and (c) the mean of the Euclidean distances (MD) between the true center c j of each physical cluster and its closest cluster representative (θ j ) obtained by each algorithm 9 . In cases where a clustering algorithm ends up with a higher number of clusters than the actual one (m f inal > m), only the m cluster representatives that are closest to the true m centers of the physical clusters, are taken into account in the determination of MD.…”

Section: The Spcm Algorithmmentioning

confidence: 99%

“…[3], [4], where each data vector is shared among two or more clusters and (c) possibilistc c-means algorithms (PCMs), e.g. [5], [6], [7], [8], [9], [1], where the compatibility of each data vector with the clusters is considered. Some significant features that both the k-means and FCM share are: (a) the interrelation of the updating equations of the representatives, (b) the requirement for a priori knowledge of the exact number of clusters m underlying in the data set, (c) the imposition of a clustering structure on the data set 1 and (d) the vulnerability to noisy data and outliers.…”

Section: Introductionmentioning

confidence: 99%

Sparsity-Aware Possibilistic Clustering Algorithms

IEEE Trans. Fuzzy Syst.

2016

In this paper two novel possibilistic clustering algorithms are presented, which utilize the concept of sparsity. The first one, called sparse possibilistic c-means, exploits sparsity and can deal well with closely located clusters that may also be of significantly different densities. The second one, called sparse adaptive possibilistic c-means, is an extension of the first, where now the involved parameters are dynamically adapted. The latter can deal well with even more challenging cases, where, in addition to the above, clusters may be of significantly different variances. More specifically, it provides improved estimates of the cluster representatives, while, in addition, it has the ability to estimate the actual number of clusters, given an overestimate of it. Extensive experimental results on both synthetic and real data sets support the previous statements.

“…However, more than one representative may be moved to the same dense region [6], [7]. Some PCM variants that try to address the problems of the conventional PCM have been reported in [8], [9], [7]. Also in [10], [11] two variations of possibilistic clustering that impose sparsity constraints, adopting the l1 norm, are proposed.…”

Section: Introductionmentioning

confidence: 99%

Sparse adaptive possibilistic clustering

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

2014

In this paper a new sparse adaptive possibilistic clustering algorithm is presented. The algorithm exhibits high immunity to outliers and provides improved estimates of the cluster representatives by adjusting dynamically certain critical parameters. In addition, the proposed scheme manages -in principle -to estimate the actual number of clusters and by properly imposing sparsity, it becomes capable to deal well with closely located clusters of different densities. Extensive experimental results verify the previous statements.