Semantic Web Services (SWS) introduce a semantic layer to the current web infrastructure, enabling the automated processing of web tasks. Different frameworks have been proposed for designing SWS, however, there has been little research in the area of testing SWS. Generating test cases for SWS is challenging due to its dynamic nature and abstract views, evaluating the test cases is equally essential in order to ensure the quality of the test suite. In this paper, we propose a goal oriented generation and mutation based evaluation approach towards SWS testing. This way, the generation (model-based) and evaluation (code-based) procedures are totally independent to each other, which further provides a much accurate measurement on the results. In addition, a tool support has been developed to automate the major steps in the proposed solution.

It is well known that the distance metric plays an important role in the clustering process. Indeed, many clustering problems can be treated as an optimization problem of a criterion function defined over one distance metric. While many distance metrics have been developed, it is not clear that how these distance metrics can impact on the clustering/optimization process. To that end, in this paper, we study the impact of a set of popular cosine-based distance metrics on K-means clustering. Specifically, by revealing the common order-preserving property, we first show that K-means has exactly the same cluster assignment for these metrics during the E-step. Next, by both theoretical and empirical studies, we prove that the cluster centroid is a good approximator of their respective optimal centers in the M-step. As such, we identify a problem with K-means: it cannot differentiate these metrics. To explore the nature of these metrics, we propose an evolutionary K-means framework that integrates K-means and genetic algorithms. This framework not only enables inspection of arbitrary distance metrics, but also can be used to investigate different formulations of the optimization problem. Finally, this framework is used in extensive experiments on real-world data sets. The results validate our theoretical findings on the characteristics and interrelationships of these metrics. Most importantly, this paper furthers our understanding of the impact of the distance metrics on the optimization process of K-means.

It is often necessary to combine multiple scores into a joint decision in many real-world applications. To that end, an essential challenge is to build a proper weighted scoring rule, which assigns more weight to the more important scores and derives

This paper proposes a Realized Trading Volatility (RTV) model for dynamically monitoring anomalous volatility in stock trading. Specifically, the RTV model first extracts the sequences for price volatility, volume volatility, and realized trading volatility. Then, the K-means algorithm is exploited for clustering the summary data of different stocks. The RTV model investigates the joint-volatility between share price and trading volume, and has the advantage of capturing anomalous trading volatility in a dynamic fashion. As a case study, we apply the RTV model for the analysis of real-world high-frequency stock data. For the resultant clusters, we focus on the categories with large volatility and study their statistical properties. Finally, we provide some empirical insights for the use of the RTV model.

Conventional k-means only considers pair wise similarity during cluster assignment, which aims to minimizing the distance of points to their nearest cluster centroids. In high dimensional space like document datasets, however, two points may be nearest neighbors without belonging to the same class. Thus pair wise similarity alone is often insufficient for class prediction in such space. To that end, in this paper, we propose to augment k-means with pair wise constraints generated from group similarity-based hyper clique patterns, which consist of strongly affiliated objects and serve as more reliable seeds for classification. Experiments with real-world datasets show that, with such constraints from quality hyper clique patterns, we can improve the clustering results in terms of various external criteria. Also, our experiments indicate that even if few constraints are violated in the original result of k-means, imposing many quality constraints may still bring gain of performance.

Ranking has been widely used in many applications. A ranking scheme usually employs a scoring rule that assigns a final numerical value to every object to be ranked. A scoring rule normally involves the use of one to many scores, and it gives more weight to the scores that are more important. In this paper, we give a scheme that can combine weights into scores in a natural way and compare our scheme to the formula given by Fagin. Also given are some additional properties that are desirable for weighted scoring rules. Finally, we discuss other interesting issues on weighted scoring rules.

Hypergraph partitioning has been considered as a promising method to address the challenges of high dimensionality in document clustering. With documents modeled as vertices and the relationship among documents captured by the hyperedges, the goal of graph partitioning is to minimize the edge cut. Therefore, the definition of hyperedges is vital to the clustering performance. While several definitions of hyperedges have been proposed, a systematic understanding of desired characteristics of hyperedges is still missing. To that end, in this paper, we first provide a unified clique perspective of the definition of hyperedges, which serves as a guide to define hyperedges. With this perspective, based on the concepts of hypercliques and shared (reverse) nearest neighbors, we propose three new types of clique hyperedges and analyze their properties regarding purity and size issues. Finally, we present an extensive evaluation using real-world document datasets. The experimental results show that, with shared (reverse) nearest neighbor based hyperedges, the clustering performance can be improved significantly in terms of various external validation measures without the need for fine tuning of parameters.

The claimed advantage of describing a document data set with a bipartite graph is that partitioning such a graph yields a co-clustering of words and documents. The topic of each cluster can then be represented by the top words and documents that have highest within-cluster degrees. However, such claims may fail if top words and documents are selected simply because they are very general and frequent. In addition, for those words and documents across several topics, it may not be proper to assign them to a single cluster. To that end, this paper introduces a new bipartite formulation that incorporates both word hypercliques and document hypercliques as super vertices. By co-preserving hyperclique patterns during the clustering process, our experiments on real-world data sets show that better clustering results can be obtained and the cluster topic can be more precisely identified. Also, we illustrate an application of the partitioned bipartite to search engines, returning clustered search results for keyword queries. We show that the topic of each cluster with respect to the current query can be identified more accurately with the words and documents from the patterns than with those top ones from the standard bipartite formulation.

The use of frequent itemsets has been limited by the high computational cost as well as the large number of resulting itemsets. In many real-world scenarios, however, it is often sufficient to mine a small representative subset of frequent itemsets with low computational cost. To that end, in this paper, we define a new problem of finding the frequent itemsets with a maximum length and present a novel algorithm to solve this problem. Indeed, maximum length frequent itemsets can be efficiently identified in very large data sets and are useful in many application domains. Our algorithm generates the maximum length frequent itemsets by adapting a pattern fragment growth methodology based on the FP-tree structure. Also, a number of optimization techniques have been exploited to prune the search space. Our experimental results show that our algorithm is very efficient and scalable

This paper describes a new bipartite formulation for word-document co-clustering such that hyperclique patterns, strongly affiliated documents in this case, are guaranteed not to be split into different clusters. Our approach for pattern preserving clustering consists of three steps: mine maximal hyperclique patterns, form the bipartite, and partition it. With hyperclique patterns of documents preserved, the topic of each cluster can be represented by both the top words from that cluster and the documents in the patterns, which are expected to be more compact and representative than those in the standard bipartite formulation. Experiments with real-world datasets show that, with hyperclique patterns as starting points, we can improve the clustering results in terms of various external clustering criteria. Also, the partitioned bipartite with preserved topical sets of documents naturally lends itself to different functions in search engines

Consensus clustering refers to combining multiple clusterings over a common set of objects into a single consolidated partition. After introducing the distribution-based view of partitions, we propose a series of entropy-based distance functions for comparing various partitions. Given a candidate partition set, consensus clustering is then formalized as an optimization problem of searching for a centroid partition with the smallest distance to that set. In addition to directly selecting the local centroid candidate, we also present two combining methods for the global centroid based on the new similarity determined by the whole candidate set. The centroid partition is likely to be top/middle-ranked in terms of closeness to the true partition. Finally we evaluate its effectiveness on both artificial and real datasets, with candidates from either the full space or the subspace.

Spatial prediction needs to account for spatial information, which makes conventional radial basis function (RBF) networks inappropriate, for they assume independent and identical distribution. In this paper, we fuse spatial information at different layers of RBF. Experiments show fusion at hidden layer gives the best result and suggest that the optimal value is around one for the coefficient, which is used in the linear combination at the output layer.