An Overview of Unsupervised and SemiSupervised Fuzzy Kernel Clustering
 Author: Frigui Hichem, Bchir Ouiem, Baili Naouel
 Organization: Frigui Hichem; Bchir Ouiem; Baili Naouel
 Publish: International Journal of Fuzzy Logic and Intelligent Systems Volume 13, Issue4, p254~268, 25 Dec 2013

ABSTRACT
For realworld clustering tasks, the input data is typically not easily separable due to the highly complex data structure or when clusters vary in size, density and shape. Kernelbased clustering has proven to be an effective approach to partition such data. In this paper, we provide an overview of several fuzzy kernel clustering algorithms. We focus on methods that optimize an fuzzy Cmeantype objective function. We highlight the advantages and disadvantages of each method. In addition to the completely unsupervised algorithms, we also provide an overview of some semisupervised fuzzy kernel clustering algorithms. These algorithms use partial supervision information to guide the optimization process and avoid local minima. We also provide an overview of the different approaches that have been used to extend kernel clustering to handle very large data sets.

KEYWORD
Fuzzy clustering , Kernelbased clustering , Relational Kernel clustering , Multiple Kernel clustering , Semisupervised clustering

1. Introduction
Clustering is an essential and frequently performed task in pattern recognition and data mining. It aids in a variety of tasks related to understanding and exploring the structure of large and high dimensional data. The goal of cluster analysis is to find natural groupings in a set of objects such that objects in the same cluster are as similar as possible and objects in different clusters are as dissimilar as possible.
In most applications, categories are rarely well separated and boundaries are overlapping. Describing these real world situations by crisp sets does not allow the user to quantitatively distinguish between objects which are strongly associated with a particular category from those that have only a marginal association with multiple ones, particularly, along the overlapping boundaries. Fuzzy clustering methods are good at dealing with these situations. In fact, data elements can belong to more than one cluster with fuzzy membership degrees.
Fuzzy clustering is widely used in the machine learning field. Areas of application of fuzzy cluster analysis include data analysis [1, 2], information retrieval [3, 4], image segmentation [5], and robotics [6]. One of the most widely used fuzzy clustering algorithm is the fuzzy Cmeans (FCM) algorithm [7].
Typically, the data to be clustered could have an object based or a relational based representation. In object data representation, each object is represented by a feature vector, while for the relational representation only information about how two objects are related is available. Relational data representation is more general in the sense that it can be applied when only the degree of dissimilarity between objects is available, or when groups of similar objects cannot be represented efficiently by a single prototype.
Despite the large number of existing clustering methods, clustering remains a challenging task when the structure of the data does not correspond to easily separable categories, and when clusters vary in size, density, and shape. Kernel based approaches [814] can adapt a specific distance measure in order to make the problem easier. They have attracted great attention and have been applied in many fields such as fault diagnosis of marine diesel engine [12], bridge parameters estimation [13], watermarking [14], automatic classiffication fragments of ceramic cultural relics [15], image segmentation [16], model construction for an erythromycin fermentation process [17], oilimmersed transformer fault diagnosis [18], analyzing enterprises independent innovation capability in order to promote different strategies [19], segmentingmagnetic resonance imaging brain images [20, 21], and classification of audio signals [22].
Kernel clustering approaches rely on their ability to produce nonlinear separating hyper surfaces between clusters by performing a mapping
ϕ from the input spaceX to a high dimensional feature spaceF . One of the most relevant aspects in kernel applications is that it is possible to compute Euclidean distances inF without knowing it explicitly. This can be done using the distance kernel trick [23]:In Eq. (1),
K (x_{i}, x_{j}) = Φ (x_{i})^{T} Φ (x_{j}) is the Mercer Kernel [24]. Gaussian kernels,are the most commonly used ones.
In this paper, we survey existing fuzzy kernel clustering algorithms. We provide an overview of unsupervised algorithms as well as semisupervised algorithms that integrate partial supervision into the objective function to guide the optimization process. For most algorithms, we describe the objective function being optimized and the necessary conditions to optimize it. We also highlight the advantages and disadvantages of the different methods.
2. Unsupervised FuzzyKernel Based Clustering
Let {x_{1}, ··· , x_{N}} be a set of
N data points to be partitioned intoC clusters, andR = [r_{jk}] is a relational matrix where r_{jk} represents the degree to which pairs of objects x_{j} and x_{k} are related. The matrixR could be given or it could be constructed from the features of the objects. Each object x_{j} belongs to every clusteri with a fuzzy membershipu_{ij} that satisfies [7]:0 ≤
u_{ij} ≤ 1,and
The exponent
m ∈ (1, ∞) is a constant that determines the level of cluster fuzziness.2.1 The Feature Space Kernel (FSK) FCM Algorithm
The FSKFCM algorithm [25] derives a kernel version of the FCM in the feature space by minimizing
subject to Eq. (3). In Eq. (4), is the center of cluster
i , in the feature space, defined as:Minimization of Eq. (4) with respect to
u_{ij} yields [25]where
The FSKFCM algorithm is one of the early kernel versions of the FCM. It is simple and has the flexibility of incorporating different kernels. It achieves this by simply fixing the update equation for the centers in the feature space. Thus, since this equation is not derived to optimize the objective function, there is no guarantee that the obtained centers are optimal. Morevover, in [25] the authors use Gaussian kernels with one global scale (𝜎) for the entire data. The selection of this parameter is not discussed in [25].
2.2 The Metric Kernel (MK) FCM Algorithm
The MKFCM [26] is an extension of the FSKFCM that allows the cluster centers to be optimized. It minimizes
subject to the constraint in Eq. (3). In Eq. (8),
ϕ (a_{i}) is the center of clusteri in the feature spaceF . Minimization of Eq. (8) has been proposed only for the case of a Gaussian kernel using the fact thatIn this case, it can be shown [26] that the update equations for the memberships and centers are
and
Unlike the FSKFCM, the MKFCM learns the optimal cluster centers in the feature space. However, the update equations have been derived only for the case of Gaussian kernels with one fixed global scale.
2.3 The Kernelized NonEuclidean Relational FCM (kNERF) Algorithm
The FSKFCM and MKFCM are object based algorithms and require an explicit feature representation of the data to be clustered. The kNERF algorithm [27], on the other hand, is a kernelized version of the nonEuclidean relational FCM algorithm [28], and works on relational data. kKERF does not formulate a new objective function. It simply uses a Gaussian kernel to compute a relational similarity matrix
R = [r _{jk}] usingThen, it uses the nonEuclidean relational fuzzy (NERF) Cmeans [28] to cluster .
kNERF is not a true kernel clustering algorithm. It simply uses kernels as a preprocessing step to create the similarity matrix. Since this step is not part of the optimization process, any kernel function can be used. However, this also prevents the kernel parameters from being optimized for the given data. Also, kNERF constructs a relational matrix with one global Gaussian parameter for the entire data. The selection of this parameter is discussed in [27] but there has been no attempt to devise methods to automatically select it.
2.4 The Clustering and Local Scale Learning (LSL) Algorithm
Although good results were obtained using the Gaussian kernel function, its performance depends on the selection of the scale parameter. Moreover, since one global 𝜎 is used for the entire data set, it may not be possible to find one optimal parameter when there are large variations between the distributions of the different clusters in the feature space. One way to learn optimal Gaussian parameters is through an exhaustive search of one parameter for each cluster. However, this approach is not practical since it is computationally expensive especially when the data include a large number of clusters and when the dynamic range of possible values for these parameters is large. Moreover, it is not trivial to evaluate the resulting partition in order to select the optimal parameters. To overcome this limitation, the LSL algorithm [29] has been proposed. It minimizes one objective function for both the optimal partition and for cluster dependent Gaussian parameters that reflect the intracluster characteristics of the data. The LSL algorithm minimizes
subject to the membership constraint in Eq. (3). The first term in Eq. (13) seeks compact clusters using a local relational distance, , with respect to each cluster
i . This distance is defined aswhere the scaling 𝜎_{i} controls the rate of decay of as a function of the distance between x_{j} and x_{k} with respect to cluster
i . The cluster dependent 𝜎_{i} allows LSL to deal with the large variations, in the feature space, between the distributions and the geometric characteristics of the different clusters. The second term in Eq. (13) is a regularization term to avoid the trivial solution where all the scaling parameters 𝜎_{i} are infinitely large.Optimization of
J with respect tou_{ij} and 𝜎_{i} yields the following update equations [29]:and
where 𝒩 is the cardinality of the neighborhood of
j .In Eq. (16), 𝜎_{i} is inversely proportional to the intracluster distances with respect to cluster
i . Thus, when the intracluster dissimilarity is small, 𝜎_{i} is large allowing the pairwise distances over the same cluster to be smaller and thus, obtain a more compact cluster. On the other hand, when the intracluster dissimilarity is high, 𝜎_{i} is small to prevent points which are not highly similar from being mapped to the same location. According to Eq. (16), 𝜎_{i} can also be seen as the average time to move between points in clusteri .LSL has the advantages of learning cluster dependent resolution parameters and can be used to idenify clusters of various densities. However, this also makes the optimization process more complex and prone to local minima. Moreover, the partition generated by LSL depends on the choice of the constant
K in Eq. (13).2.5 The Fuzzy ClusteringWith Learnable Cluster Dependent Kernels (FLeCK) Algorithm
FLeCK [30] is an extension of LSL that does not assume that
K is fixed. It learns the scaling parameters andK by optimizing both the intracluster and the intercluster dissimilarities. Consequently, the learned scale parameters reflect the relative density, size, and position of each cluster with respect to the other clusters. In particular, FLeCK minimizes the intracluster distancesand maximizes the intercluster distances
The constant
K provides a way of specifying the relative importance of the regularization term (second term in Eqs. (17) and 18)), compared to the sum of intracluster (first term in Eq. (17)) and intercluster distances (first term in Eq. (18)). The parameter, 𝜎_{i}, controls the rate of decay of . This approach learns a cluster dependent 𝜎_{i} in order to deal with the large variations between the distributions and the geometric characteristics of each cluster.Using the Lagrange multiplier technique and assuming that the values of 𝜎_{i} do not change significantly from one iteration (
t − 1) to the next one (t ), it can be shown [30] thatwhere
and
The simultaneous optimization of Eqs. (17) and (18) with respect to
u_{ij} yields [30]:where
and is as defined in Eq. (14).
2.6 The Fuzzy ClusteringWith MultipleKernels (FCMK) Algorithm
LSL [29] and FLeCK [30] have the advantage of learning cluster dependent scaling parameters. Thus, they can be used to identify clusters of different densities. However, these algorithms have two main limitations. First, learning 𝜎_{i} for each cluster involves complex optimization and is prone to local minima. Second, data points assigned to one cluster are constrained to have one common density. An alternative approach, that involves relatively simpler optimization, and overcomes the above limitations is based on multiple kernel learning (MKL) [3134]. Instead of using one kernel (fixed or learned) per cluster, MKL uses multiple kernels for each cluster.
Most existing MKL methods assume that kernel weights remain constant for all data (i.e., spacelevel), while algorithms like Localized MKL [35] seek kernel weights that are data dependent and locally smooth (i.e., samplelevel). Although samplelevel nonuniform methods give the largest flexibility, in practice relaxations are typically introduced to enhance tractability. Most of the previous MKL approaches have focused on supervised [32, 33] and semisupervised learning [35]. Recently, an extension of multiple kernels to unsupervised learning, based on maximum margin clustering, was proposed in [36]. However, this method is only designed for crisp clustering. In [37], Huang et al. designed a multiple kernel fuzzy clustering algorithm which uses one set of global kernel weights for all clusters. Therefore, their approach is not appropriate for clusters of various densities. To overcome these drawbacks, Baili and Frigui proposed the FCMK [38]. FCMK is based on the observation that data within the same cluster tend to manifest similar properties. Thus, the intracluster weights can be approximately uniform while kernel weights are allowed to vary across clusters.
Given a set of
M embedding mappings that map the data to new feature spaces, Φ = {ϕ 1, . . . ,ϕ_{M} }. Each mappingϕ_{k} maps thep dimensional datax as a vectorϕ_{k} (x ) in its feature space whose dimensionality isL_{k} . Let {K _{1}, . . . ,K_{M} } be the kernels corresponding to these implicit mapping respectively,Since these implicit mappings do not necessarily have the same dimensionality, the authors in [38] construct a new set of independent mappings, Ψ = ｛
ψ _{1};ψ _{2}; ... ;ψ _{M}｝, from the original mappings Φ asEach of these constructed mappings converts
x into anL  dimensional vector, where . The linear combination of the bases in Ψ within each clusteri is defined aswith
A new clusterdependent kernel,
K ^{(i)}, between objectj and clusteri , is computed as a convex combination ofM Gaussian kernelsK _{1},K _{2}, . . . ,K_{M} with fixed scaling parameters 𝜎_{1}, 𝜎_{2}, . . . , 𝜎_{M}, respectively. That is,In Eq. (26), W = [
w_{ik} ], wherew_{ik} ∈ [0; 1] is a resolution weight for kernelK_{k} with respect to clusteri . A low value ofw_{ik} indicates that kernelK_{k} is not relevant for the density estimation of clusteri (due to its scaling parameter), and that this kernel should not have a significant impact on the creation of this cluster. Similarly, a high value ofw_{ik} indicates that the bandwidth of kernelK_{k} is highly relevant for the density estimation of clusteri , and that this kernel should be the main factor in the creation of this cluster.The FCMK algorithm minimizes
subject to the constraints in Eqs. (3) and (25). It can be shown [38] that minimization of Eq. (27) with respect to
u_{ij} yieldswhere
D_{ij} denotes the distance between datax_{j} and cluster centera_{i} , i.e., . Similarly, minimization of Eq. (27) with respect toa_{i} yieldswhere is the normalized membership.
Using Eq. (29), the distance between data
x_{j} and cluster centera_{i} can be written aswhere the coefficients
α_{ijk} are defined asReplacing Eq. (30) back in Eq. (27) and introducing a Lagrange multiplier, the optimal kernel weights need to be updated using
where
The FCMK is based on the MKFCM [26] and inherits the limitations of objectbased clustering. First, multiple runs of the algorithm may be needed since it is susceptible to initialization problems. Second, FCMK is restricted to data for which there is a notion of center (centroid). Finally, FCMK is not suitable for all types of data. For instance, the density is estimated by counting the number of points within a specified radius, 𝜎_{k}, of the cluster center. However, for data with multiple resolutions within the same cluster, density should take into account variance between pairs of points and not points to center.
2.7 The Relational Fuzzy Clustering With Multiple Kernels (RFCMK) Algorithm
To overcome the limitations of FCMK, the RFCMK was proposed in [39]. The RFCMK involves dissimilarity between pairs of objects instead of dissimilarity of the objects to a cluster center. It minimizes
subject to the constraints in Eqs. (3) and (25). In Eq. (34), is the transformed relational data between feature points
x_{j} andx_{h} , with respect to clusteri define using the implicit mappingψ ^{(i)}:The distance from feature vector
x_{j} to the center of thei^{th} cluster,a_{i} , can be written in terms of the relational matrix using [40]where υ_{i} is the membership vector defined by
It can be shown [39] that optimiation of Eq. (34) with respect to
u_{ij} yieldsRewriting the relational data Eq. (35) as
where
it can be shown [39] that the kernel weights need to be updated using
where
In Eq. (41), the resolution weight
w_{ik} is inversely proportional toα_{jhk} , which is the distance between objectsx_{j} andx_{h} induced by kernelk . When, the objects are mapped close to each other,w_{ik} will be large indicating that kernelk is relevant. On the other hand, when, the objects are mapped far apart, the weightw_{ik} will be small indicating that kernelk is irrelevant.3. SemiSupervised Fuzzy Kernel Based Clustering
Clustering is a difficult combinatorial problem that is susceptible to local minima, especially for high dimensional data. Incorporating prior knowledge in the unsupervised learning task, in order to guide the clustering process has attracted considerable interest among researchers in the data mining and machine learning communities. Some of these methods use available sideinformation to learn the parameters of the Mahalanobis distance (e.g. [41–43]). The Nonlinear methods have focused on the kernelization of Mahalanobis metric learning methods (e.g. [44–46]). However, these approaches are computationally expensive or even infeasible for high dimensional data.
Another approach to guide the clustering process is to use available information in the form of hints, constraints, or labels. Supervision in the form of constraints is more practical than providing class labels. This is because in many real world applications, the true class labels may not be known, and it is much easier to specify whether pairs of points should belong to the same or to different clusters. In fact, pairwise constraints occur naturally in many domains.
3.1 The SemiSupervised Kernel FCM (SSKFCM) Algorithm
The SSKFCM [47] uses
L labeled points andU unlabeld points. It assumes that fuzzy memberships can be assigned, using expert knowledge, to the subset of labeled data. Its objective function is defined by adding classification errors of both the labeled and the unlabeled data, i.e.,The first term in Eq. (43) is the error between the fuzzy centers calculated based on the learned clusters and the labeled information. The second term minimizes the intracluster distances. The optimal solutions to Eq. (43) involves learning the fuzzy memberships and the kernel parameters. It can be shown [47] that for the labeled data, the memberships need to be updated using:
u_{ij} =and for the unlabeled data, the fuzzy membership need to be updated using:
Optimization of Eq. (43) with respect to 𝜎 does not lead to a closed form expression. Instead, 𝜎 is updated iteratively using:
The SSKFCM algorithm assumes that a subset of the data is labeled with fuzzy membership degrees. However, for real applications, this information may not be available and can be tedious to generate using expert knowledge. An alternative approach uses pairwise constraints. For the following semisupervised algorithms, we assume that pairwise “
ShouldLink ” constraints (pairs of points that should belong to the same cluster) and “Should notLink ” constraints (pairs of points that should belong to different clusters) are provided with the input. letSl be the indicator matrix for the set of “ShouldLink ” pairs of constraints such thatSl (j ,k ) = 1 means that x_{j} and x_{k} should be assigned to the same cluster and 0 otherwise. Similarly, letSNl be the indicator matrix for the set of “ShouldLink ” pairs such thatSNl (j ,k ) = 1 means that x_{j} and x_{k} should not be assigned to the same cluster and 0 otherwise.3.2 SemiSupervised Relational ClusteringWith Local Scaling
The semisupervised local scaling learning (SSLSL) [48] minimizes
subject to the constraint in Eq. (3). SSLSL is an extension of the LSL algorithm [29] that incorporates partial supervision. The second term in Eq. (47), is a reward term for satisfying “
ShouldLink ” constraints and a penalty term for violating “Should notLink ” constraints.In Eq. (47), the weight
w ∈ (0; 1) provides a way of specifying the relative importance of the “ShouldLink ” and “Should notLink ” constraints compared to the sum of intercluster distances. In [48], the authors recommend fixing it as the ratio of the number of constraints to the total number of points.Setting the derivative of
J with respect to 𝜎_{i} gives the same update equation for 𝜎_{i} as the LSL algorithm [29] (Refer to Eq. (16)).The objective function in Eq. (47) can be rewritten as
where
can be regarded as the “effective distance” that takes into account the satisfaction and violation of the constraints.
It can be shown [48] that optimization of
J with respect tou_{ij} yieldswhere
3.3 The SSFLeCK Algorithm
SSFLeCK [49] is an extension of FLeCK [30] that incorporates partial supervision information. It attempts to satisfy a set of “
ShouldLink ” and “Should notLink ” constraints by minimizingThe optimal
K is determined by simultaneously minimizing the intracluster distances Eq. (52) and maximizing the intercluster distances in Eq. (18).It can be shown [49] that optimization of Eq. (52) and Eq. (18) with respect to σ_{i} yields the same update equation for 𝜎_{i} as in FLeCK (i.e., Eq. (19)). Similarly, optimization of Eq. (52) with respect to the memberships yields the same update equation as SSLSL (i.e. Eq. (50)).
3.4 The SSFCMK Algorithm
The SSFCMK [50] uses the same partial supervision information to extend FCMK [38] by minimizing:
subject to Eqs. (3) and (25).
In Eq. (53), the weight
γ ∈ (0; 1) provides a way of specifying the relative importance of theshouldlink andshould notlink constraints compared to the sum of intercluster distances. It can be shown [50] that the update equation for the membership of pointx_{j} to clusteri iswhere is given by Eq. (28), and
In Eq. (55),
C_{ij} and are defined asand
The first term in Eq. (54) is the membership term in the FCMK algorithm and only focuses on distances between feature points and prototypes. The second term takes into account the available supervision: memberships are reinforced or reduced according to the pairwise constraints provided by the user.
C_{ij} is the constraint violation cost associated with feature pointx_{j} if it is assigned to clusteri , while is the weighted average, over all the clusters, of the violation costs associated with feature pointx_{j} . If the violated constraints do not involve feature pointx_{j} , thenSince the constraints in Eq. (53) do not depend on
W explicitly, optimization of Eq. (53) with respect toW yields the same update Eq. (32) as in FCMK.3.5 The SSRFCMK Algorithm
The SSRFCMK [50] used partial supervision to extend RFCMK [39]. It minimizes
subject to Eqs. (3) and (25).
Using the Lagrange multiplier technique, it can be shown [50] that the memberships need to be updated using
where is as defined in Eq. (38)
In Eq. (60),
C_{ij} and are as defined in Eqs. (56) and (57).Since the constraints in Eq. (58) do not depend on
w_{ik} explicitly, optimization of Eq. (58) with respect toW yields the same update Eq. (41) as in RFCMK.4. Other Fuzzy Kernel Clustering Algorithms
In this paper, we have focused on kernel clustering algorithms that are based on the FCM objective function. There are several other fuzzy kernel clustering approaches. For instance, the multisphere support vectors (MSV) clustering [51] extends the SV clustering approach [52]. It defines a cluster as a sphere in the feature space. This yields kernelbased mapping between the original space and the feature space. The kernel possibilistic Cmeans (KPCM) algorithm [53] applies the kernel approach to the possibilistic Cmeans (PCM) algorithm [54]. The weighted kernel fuzzy clustering algorithm (WKFCA) [55] is a kernelized version of the SCAD algorithm [56]. It performs feature weighting along with fuzzy kernel clustering. In [57], the authors propose a kernel fuzzy clustering model that extends the additive fuzzy clustering model to the case of a nonlinear model. More specifically, it has been shown that the additive clustering [58] is special case of fuzzy kernel clustering. The similarity structure is then captured and mapped to higher dimensional space by using kernel functions. The kernel fuzzy clustering methods based on local adaptive distances [59] performs feature weighting and fuzzy kernel clustering simultaneously. The sum of the weights of the variables with respect to each cluster is not equal to one as in [55]. However, the product of the weights of the variables for each cluster is constrained be equal to one. The genetic multiple kernel interval type 2 FCM clustering [60] combines heuristic method based on genetic algorithm (GA) and MKFCM. It automatically determines the optimal number of clusters and the initial centroids in the first step. Then, it adjusts the coefficients of the kernels and combines them in the feature space to produce a new kernel. Other kernel clustering algorithms, based on type 2 fuzzy sets, include [61, 62]. A kernel intuitionistic FCM clustering algorithm (KIFCM) was proposed in [63]. EKIFCM has two main phases. The first one is KIFCM and the second phase is parameters selection of KIFCM with GA. KIFCM is a combination of Atanassov's intuitionistic fuzzy sets (IFSs) [64] with kernelbased FCM (KFCM) [47].
5. FuzzyKernel Based Clustering for Very Large Data
All of the kernel clustering algorithms that we have outlined do not scale to very large (VL) data sets. VL data or big data are any data that cannot be loaded into the computer's working memory. Since clustering is one of the primary tasks used in the pattern recognition and data mining communities to search VL databases in various applications, it is desirable to have clustering algorithms that scale well to VL data.
The scalability issue has been studied by many researchers. In general, there are three main approaches to clustering for VL data: samplingbased methods, incremental algorithms and data approximation algorithms. The sample and extend algorithms apply clustering to a sample of the dataset found by either progressive [65] or random sampling [66]. Then, the sample results are noniteratively extended to approximate the partition for the rest of the data. Representative algorithms include random sample and extend kernel FCM (rseKFCM) algorithm [67] and the random and extend RFCMK (rseRFCMK) [68]. The rseRFCMK is an extension of the RFCMK algorithm to VL data based on sampling followed by noniterative extension. The main problem with samplingbased methods is the choice of the sample. For instance, if the sample is not representative of the full dataset, then sample and extend methods cannot accurately approximate the clustering solution.
On the other hand, the incremental algorithms are designed to operate on the full dataset by separating it into multiple chunks. First, these algorithms sequentially load manageable chunks of the data. Then, they cluster each chunk in a single pass. They construct the final partition as a combination of the results from each chunk. In [66, 67], a single pass kernel fuzzy Cmeans (spKFCM) algorithm was proposed. The spKFCM algorithm runs weighted KFCM (wKFCM) on sequential chunks of the data, passing the clustering solution from each chunk onto the next. spKFCM is scalable as its space complexity is only based on the size of the sample. The single pass RFCMK (spRFCMK) [68] is an extension of the RFCMK algorithm to VL data. The spRFCMK is an incremental technique that makes one sequential pass through subsets of the data.
In [66], an online algorithm for kernel FCM (oKFCM) was proposed in which data is assumed to arrive in chunks. Each chunk is clustered and the memory is freed by summarizing the clustering result by weighted centroids. In contrast to spKFCM, oKFCM is not truly scalable and is not recommended for VL data. This is because, rather than passing the clustering results from one chunk to the next, oKFCM clusters the weighted centroids in one final run to obtain the clustering for the entire stream.
6. Conclusion
Fuzzy kernel clustering has proven to be an effective approach to partition data when the clusters are not welldefined. Several algorithms have been proposed in the past few years and were described in this paper. Some algorithms, such as FSKFCM, are simple and can incorporate any kernel function. However, these methods impose an intuitive equation to update the centers in the feature space. Thus, there is no guarantee that the optimized objective function of these algorithms correspond to the optimal partition.
Other kernel algorithms can solve for the optimal centers by restricting the kernel to be Gaussian. Some of these algorithms, such as FSKFCM, use one global scale (𝜎) for all clusters. These are relatively simpler algorithms that are not very sensitive to the initialization. However, they require the user to specify the scale, and may not perform well when the data exhibit large variations between the distributions of the different clusters. Other algorithms, such as FLeCK, use more complex objective functions to learn clusterdependent kernel resolution parameters. However, because the search space of these methods include many more parameters, they are more prone to local minima. Clustering with multiple kernels is a good compromize between methods that use one fixed global scale and methods that learn one scale for each cluster. These algorithms, such as FCMK, use a set of kernels with different, but fixed, scales. Then, they learn relevance weights for each kernel within each cluster.
Clustering, in general, is a difficult combinatorial problem that is susceptible to local minima. This problem is more acute in Kernel based clustering as they solve the partitioning problem in a much higher feature space. As a result, several semisupervised kernel clustering methods have emerged. These algorithms incorporate prior knowledge in order to guide the optimization process. This prior knowledge is usually available in the form of a small set of constraints that specify which pairs of points should be assigned to the same cluster, and which ones should be assigned to different clusters.
Scalability to very large data is another desirable feature to have in a clustering algorithm. In fact, many applications involve very large data that cannot be loaded into the computer's memory. In this case, algorithm scalability becomes a necessary condition. We have outlined three main approaches that have been used with kernel clustering methods. These include samplingbased methods, incremental algorithms, and data approximation algorithms.
Conflict of Interest
No potential conflict of interest relevant to this article was reported.

18. D. H. Liu, Bian J. P., Sun X. Y. 2008 “The study of fault diagnosis model of DGA for oilimmersed transformer based on fuzzy means Kernel clustering and SVM multiclass object simplified structure,” [Proceedings of the 7th International Conference on Machine Learning and Cybernetics] P.15051509

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]