Evaluation of the performance of clustering algorithms in kernel-induced feature space

By using a kernel function, data that are not easily separable in the original space can be clustered into homogeneous groups in the implicitly transformed high-dimensional feature space. Kernel k-means algorithms have recently been shown to perform better than conventional k-means algorithms in unsupervised classification. However, few reports have examined the benefits of using a kernel function and the relative merits of the various kernel clustering algorithms with regard to the data distribution. In this study, we reformulated four representative clustering algorithms based on a kernel function and evaluated their performances for various data sets. The results indicate that each kernel clustering algorithm gives markedly better performance than its conventional counterpart for almost all data sets. Of the kernel clustering algorithms studied in the present work, the kernel average linkage algorithm gives the most accurate clustering results.     

Hybrid differential evolution algorithm for optimal clustering

The problem of optimal non-hierarchical clustering is addressed. A new algorithm combining differential evolution and k-means is proposed and tested on eight well-known real-world data sets. Two criteria (clustering validity indexes), namely TRW and VCR, were used in the optimization of classification. The classification of objects to be optimized is encoded by the cluster centers in differential evolution (DE) algorithm. It induced the problem of rearrangement of centers in the population to ensure an efficient search via application of evolutionary operators. A new efficient heuristic for this rearrangement was also proposed. The plain DE variants with and without the rearrangement were compared with corresponding hybrid k-means variants. The experimental results showed that hybrid variants with k-means algorithm are essentially more efficient than the non-hybrid ones. Compared to a standard k-means algorithm with restart, the new hybrid algorithm was found more reliable and more efficient, especially in difficult tasks. The results for TRW and VCR criterion were compared. Both criteria provided the same optimal partitions and no significant differences were found in efficiency of the algorithms using these criteria.     

A hybrid MapReduce-based <Emphasis Type="Italic">k</Emphasis>-means clustering using genetic algorithm for distributed datasets

Clustering a large volume of data in a distributed environment is a challenging issue. Data stored across multiple machines are huge in size, and solution space is large. Genetic algorithm deals effectively with larger solution space and provides better solution. In this paper, we proposed a novel clustering algorithm for distributed datasets, using combination of genetic algorithm (GA) with Mahalanobis distance and k-means clustering algorithm. The proposed algorithm is two phased; in phase 1, GA is applied in parallel on data chunks located across different machines. Mahalanobis distance is used as fitness value in GA, which considers covariance between the data points and thus provides a better representation of initial data. K-means with K-means\( ++ \) initialization is applied in phase 2 on intermediate output to get final result. The proposed algorithm is implemented on Hadoop framework, which is inherently designed to deal with distributed datasets in a fault-tolerant manner. Extensive experiments were conducted for multiple real-life and synthetic datasets to measure performance of our proposed algorithm. Results were compared with MapReduce-based algorithms, mrk-means, parallel k-means and scaling GA.     

Semi-supervised graph clustering: a kernel approach

Semi-supervised clustering algorithms aim to improve clustering results using limited supervision. The supervision is generally given as pairwise constraints; such constraints are natural for graphs, yet most semi-supervised clustering algorithms are designed for data represented as vectors. In this paper, we unify vector-based and graph-based approaches. We first show that a recently-proposed objective function for semi-supervised clustering based on Hidden Markov Random Fields, with squared Euclidean distance and a certain class of constraint penalty functions, can be expressed as a special case of the weighted kernel k-means objective (Dhillon et al., in Proceedings of the 10th International Conference on Knowledge Discovery and Data Mining, 2004a). A recent theoretical connection between weighted kernel k-means and several graph clustering objectives enables us to perform semi-supervised clustering of data given either as vectors or as a graph. For graph data, this result leads to algorithms for optimizing several new semi-supervised graph clustering objectives. For vector data, the kernel approach also enables us to find clusters with non-linear boundaries in the input data space. Furthermore, we show that recent work on spectral learning (Kamvar et al., in Proceedings of the 17th International Joint Conference on Artificial Intelligence, 2003) may be viewed as a special case of our formulation. We empirically show that our algorithm is able to outperform current state-of-the-art semi-supervised algorithms on both vector-based and graph-based data sets.     

Fast global k-means clustering using cluster membership and inequality

In this paper, we present a fast global k-means clustering algorithm by making use of the cluster membership and geometrical information of a data point. This algorithm is referred to as MFGKM. The algorithm uses a set of inequalities developed in this paper to determine a starting point for the jth cluster center of global k-means clustering. Adopting multiple cluster center selection (MCS) for MFGKM, we also develop another clustering algorithm called MFGKM+MCS. MCS determines more than one starting point for each step of cluster split; while the available fast and modified global k-means clustering algorithms select one starting point for each cluster split. Our proposed method MFGKM can obtain the least distortion; while MFGKM+MCS may give the least computing time. Compared to the modified global k-means clustering algorithm, our method MFGKM can reduce the computing time and number of distance calculations by a factor of 3.78-5.55 and 21.13-31.41, respectively, with the average distortion reduction of 5,487 for the Statlog data set. Compared to the fast global k-means clustering algorithm, our method MFGKM+MCS can reduce the computing time by a factor of 5.78-8.70 with the average reduction of distortion of 30,564 using the same data set. The performances of our proposed methods are more remarkable when a data set with higher dimension is divided into more clusters.     

A feature group weighting method for subspace clustering of high-dimensional data

This paper proposes a new method to weight subspaces in feature groups and individual features for clustering high-dimensional data. In this method, the features of high-dimensional data are divided into feature groups, based on their natural characteristics. Two types of weights are introduced to the clustering process to simultaneously identify the importance of feature groups and individual features in each cluster. A new optimization model is given to define the optimization process and a new clustering algorithm FG-k-means is proposed to optimize the optimization model. The new algorithm is an extension to k-means by adding two additional steps to automatically calculate the two types of subspace weights. A new data generation method is presented to generate high-dimensional data with clusters in subspaces of both feature groups and individual features. Experimental results on synthetic and real-life data have shown that the FG-k-means algorithm significantly outperformed four k-means type algorithms, i.e., k-means, W-k-means, LAC and EWKM in almost all experiments. The new algorithm is robust to noise and missing values which commonly exist in high-dimensional data.     

An optimal method for data clustering

An algorithm for optimizing data clustering in feature space is studied in this work. Using graph Laplacian and extreme learning machine (ELM) mapping technique, we develop an optimal weight matrix W for feature mapping. This work explicitly performs a mapping of the original data for clustering into an optimal feature space, which can further increase the separability of original data in the feature space, and the patterns points in same cluster are still closely clustered. Our method, which can be easily implemented, gets better clustering results than some popular clustering algorithms, like k-means on the original data, kernel clustering method, spectral clustering method, and ELM k-means on data include three UCI real data benchmarks (IRIS data, Wisconsin breast cancer database, and Wine database).     

Equally contributory privacy-preserving k-means clustering over vertically partitioned data

In recent years, there have been numerous attempts to extend the k-means clustering protocol for single database to a distributed multiple database setting and meanwhile keep privacy of each data site. Current solutions for (whether two or more) multiparty k-means clustering, built on one or more secure two-party computation algorithms, are not equally contributory, in other words, each party does not equally contribute to k-means clustering. This may lead a perfidious attack where a party who learns the outcome prior to other parties tells a lie of the outcome to other parties. In this paper, we present an equally contributory multiparty k-means clustering protocol for vertically partitioned data, in which each party equally contributes to k-means clustering. Our protocol is built on ElGamal's encryption scheme, Jakobsson and Juels's plaintext equivalence test protocol, and mix networks, and protects privacy in terms that each iteration of k-means clustering can be performed without revealing the intermediate values.     

Fast modified global k-means algorithm for incremental cluster construction

The k-means algorithm and its variations are known to be fast clustering algorithms. However, they are sensitive to the choice of starting points and are inefficient for solving clustering problems in large datasets. Recently, incremental approaches have been developed to resolve difficulties with the choice of starting points. The global k-means and the modified global k-means algorithms are based on such an approach. They iteratively add one cluster center at a time. Numerical experiments show that these algorithms considerably improve the k-means algorithm. However, they require storing the whole affinity matrix or computing this matrix at each iteration. This makes both algorithms time consuming and memory demanding for clustering even moderately large datasets. In this paper, a new version of the modified global k-means algorithm is proposed. We introduce an auxiliary cluster function to generate a set of starting points lying in different parts of the dataset. We exploit information gathered in previous iterations of the incremental algorithm to eliminate the need of computing or storing the whole affinity matrix and thereby to reduce computational effort and memory usage. Results of numerical experiments on six standard datasets demonstrate that the new algorithm is more efficient than the global and the modified global k-means algorithms.     

A novel incremental conceptual hierarchical text clustering method using CFu-tree

As a data mining method, clustering, which is one of the most important tools in information retrieval, organizes data based on unsupervised learning which means that it does not require any training data. But, some text clustering algorithms cannot update existing clusters incrementally and, instead, have to recompute a new clustering from scratch. In view of above, this paper presents a novel down-top incremental conceptual hierarchical text clustering approach using CFu-tree (ICHTC-CF) representation, which starts with each item as a separate cluster. Term-based feature extraction is used for summarizing a cluster in the process. The Comparison Variation measure criterion is also adopted for judging whether the closest pair of clusters can be merged or a previous cluster can be split. And, our incremental clustering method is not sensitive to the input data order. Experimental results show that the performance of our method outperforms k-means, CLIQUE, single linkage clustering and complete linkage clustering, which indicate our new technique is efficient and feasible.     

Finding groups in data: Cluster analysis with ants

We present in this paper a modification of Lumer and Faieta’s algorithm for data clustering. This approach mimics the clustering behavior observed in real ant colonies. This algorithm discovers automatically clusters in numerical data without prior knowledge of possible number of clusters. In this paper we focus on ant-based clustering algorithms, a particular kind of a swarm intelligent system, and on the effects on the final clustering by using during the classification different metrics of dissimilarity: Euclidean, Cosine, and Gower measures. Clustering with swarm-based algorithms is emerging as an alternative to more conventional clustering methods, such as e.g. k-means, etc. Among the many bio-inspired techniques, ant clustering algorithms have received special attention, especially because they still require much investigation to improve performance, stability and other key features that would make such algorithms mature tools for data mining.As a case study, this paper focus on the behavior of clustering procedures in those new approaches. The proposed algorithm and its modifications are evaluated in a number of well-known benchmark datasets. Empirical results clearly show that ant-based clustering algorithms performs well when compared to another techniques.     

Improving the performance of k-means for color quantization

Color quantization is an important operation with many applications in graphics and image processing. Most quantization methods are essentially based on data clustering algorithms. However, despite its popularity as a general purpose clustering algorithm, k-means has not received much respect in the color quantization literature because of its high computational requirements and sensitivity to initialization. In this paper, we investigate the performance of k-means as a color quantizer. We implement fast and exact variants of k-means with several initialization schemes and then compare the resulting quantizers to some of the most popular quantizers in the literature. Experiments on a diverse set of images demonstrate that an efficient implementation of k-means with an appropriate initialization strategy can in fact serve as a very effective color quantizer.     

A k-means type clustering algorithm for subspace clustering of mixed numeric and categorical datasets

Almost all subspace clustering algorithms proposed so far are designed for numeric datasets. In this paper, we present a k-means type clustering algorithm that finds clusters in data subspaces in mixed numeric and categorical datasets. In this method, we compute attributes contribution to different clusters. We propose a new cost function for a k-means type algorithm. One of the advantages of this algorithm is its complexity which is linear with respect to the number of the data points. This algorithm is also useful in describing the cluster formation in terms of attributes contribution to different clusters. The algorithm is tested on various synthetic and real datasets to show its effectiveness. The clustering results are explained by using attributes weights in the clusters. The clustering results are also compared with published results.     

Scale-invariant clustering with minimum volume ellipsoids

This paper develops theory and algorithms concerning a new metric for clustering data. The metric minimizes the total volume of clusters, where the volume of a cluster is defined as the volume of the minimum volume ellipsoid (MVE) enclosing all data points in the cluster. This metric is scale-invariant, that is, the optimal clusters are invariant under an affine transformation of the data space. We introduce the concept of outliers in the new metric and show that the proposed method of treating outliers asymptotically recovers the data distribution when the data comes from a single multivariate Gaussian distribution. Two heuristic algorithms are presented that attempt to optimize the new metric. On a series of empirical studies with Gaussian distributed simulated data, we show that volume-based clustering outperforms well-known clustering methods such as k-means, Ward's method, SOM, and model-based clustering.     

A K-partitioning algorithm for clustering large-scale spatio-textual data

The volume of spatio-textual data is drastically increasing in these days, and this makes more and more essential to process such a large-scale spatio-textual dataset. Even though numerous works have been studied for answering various kinds of spatio-textual queries, the analyzing method for spatio-textual data has rarely been considered so far. Motivated by this, this paper proposes a k-means based clustering algorithm specialized for a massive spatio-textual data. One of the strong points of the k-means algorithm lies in its efficiency and scalability, implying that it is appropriate for a large-scale data. However, it is challenging to apply the normal k-means algorithm to spatio-textual data, since each spatio-textual object has non-numeric attributes, that is, textual dimension, as well as numeric attributes, that is, spatial dimension. We address this problem by using the expected distance between a random pair of objects rather than constructing actual centroid of each cluster. Based on our experimental results, we show that the clustering quality of our algorithm is comparable to those of other k-partitioning algorithms that can process spatio-textual data, and its efficiency is superior to those competitors.     

求解大规模谱聚类的近似加权核k-means算法

谱聚类将聚类问题转化成图划分问题,是一种基于代数图论的聚类方法.在求解图划分目标函数时,一般利用Rayleigh熵的性质,通过计算Laplacian矩阵的特征向量将原始数据点映射到一个低维的特征空间中,再进行聚类.然而在谱聚类过程中,存储相似矩阵的空间复杂度是O(n²),对Laplacian矩阵特征分解的时间复杂度一般为O(n³),这样的复杂度在处理大规模数据时是无法接受的.理论证明,Normalized Cut图聚类与加权核k-means都等价于矩阵迹的最大化问题.因此,可以用加权核k-means算法来优化Normalized Cut的目标函数,这就避免了对Laplacian矩阵特征分解.不过,加权核k-means算法需要计算核矩阵,其空间复杂度依然是O(n²).为了应对这一挑战,提出近似加权核k-means算法,仅使用核矩阵的一部分来求解大数据的谱聚类问题.理论分析和实验对比表明,近似加权核k-means的聚类表现与加权核k-means算法是相似的,但是极大地减小了时间和空间复杂性.     

How to “alternatize” a clustering algorithm

Given a clustering algorithm, how can we adapt it to find multiple, nonredundant, high-quality clusterings? We focus on algorithms based on vector quantization and describe a framework for automatic ‘alternatization’ of such algorithms. Our framework works in both simultaneous and sequential learning formulations and can mine an arbitrary number of alternative clusterings. We demonstrate its applicability to various clustering algorithms—k-means, spectral clustering, constrained clustering, and co-clustering—and effectiveness in mining a variety of datasets.     

k-ANMI: A mutual information based clustering algorithm for categorical data

Clustering categorical data is an integral part of data mining and has attracted much attention recently. In this paper, we present k-ANMI, a new efficient algorithm for clustering categorical data. The k-ANMI algorithm works in a way that is similar to the popular k-means algorithm, and the goodness of clustering in each step is evaluated using a mutual information based criterion (namely, average normalized mutual information – ANMI) borrowed from cluster ensemble. This algorithm is easy to implement, requiring multiple hash tables as the only major data structure. Experimental results on real datasets show that k-ANMI algorithm is competitive with those state-of-the-art categorical data clustering algorithms with respect to clustering accuracy.     

Adaptive multilevel rough entropy evolutionary thresholding

In this study, comprehensive research into rough set entropy-based thresholding image segmentation techniques has been performed producing new and robust algorithmic schemes. Segmentation is the low-level image transformation routine that partitions an input image into distinct disjoint and homogenous regions using thresholding algorithms most often applied in practical situations, especially when there is pressing need for algorithm implementation simplicity, high segmentation quality, and robustness. Combining entropy-based thresholding with rough set results in the rough entropy thresholding algorithm.The authors propose a new algorithm based on granular multilevel rough entropy evolutionary thresholding that operates on a multilevel domain. The MRET algorithm performance has been compared to the iterative RET algorithm and standard k-means clustering methods on the basis of β-index as a representative validation measure. Performance in experimental assessment suggests that granular multilevel rough entropy threshold based segmentations - MRET - present high quality, comparable with and often better than k-means clustering based segmentations. In this context, the rough entropy evolutionary thresholding MRET algorithm is suitable for specific segmentation tasks, when seeking solutions that incorporate spatial data features with particular characteristics.