With the increasing of data and Web users in the Internet, it is necessary and significant to identify users for offering further possible personalized services. To implement this goal, we used the users' surfing historical data stored in the Web log as features for classification. This paper aims to develop some techniques to achieve this objective using fuzzy neural networks (FNNs). Experimental results demonstrate the effectiveness of our proposed approach.

Quality of Service (QoS) is an important factor for selecting a better Web service from numerous semantic equivalent Web services in Web services composition. In this paper, a QoS model (e-QoS) for evaluating semantic equivalent Web services is proposed, In which, the factors used to evaluate Web Services include Accessed Times, Invoked Success Rate, Average Responding Time, Stability Variance, and Customers Estimation. The performance of Web services in Latest Period of Time, the performance in history and the performance estimated by customers are considered together. By experiments, we concluded that e_QoS is effective for estimating Web services, and can improve the quality of web services composition greatly.

To support very high dimensionality, model-based clustering is an intuitive choice for document clustering. However, the current model-based algorithms are prone to generating the skewed clusters, which influence the quality of clustering seriously. In this paper, the reasons of skew are examined and determined as the inappropriate initial model, the unfitness of cluster model and the interaction between the decentralization of estimation samples and the over-generalized cluster model. This paper proposes a skew prevention document-clustering algorithm (MMPClust), which has two features: (1) a content-based cluster model is used to model the cluster better; (2) at the re-estimation step, a part of documents most relevant to its corresponding class are selected automatically for each cluster as the estimation samples to break this interaction. MMPClust has less restrictions and more applicability in document clustering than the previous methods.

XML is rapidly emerging as the dominant standard for exchanging data on the WWW. Most of application data are stored in relational databases due to its popularity and rich development experiences over it. Therefore, how to provide a proper mapping approach from relational data to XML documents becomes an important topic. Integrity constraints are useful for semantic specification that plays the important roles in relation schema definition. The existing XML schema language does not define general constraints and maintaining method for integrity constraints. So how to use XML to express and maintain integrity constraints especially the advanced integrity constraints, e.g., general constraints of relational data is one of challenge research issues. In this paper, a novel mapping approach is proposed to map relation data to XML document with active nodes, XMLA, and extended DTD with constraints, DTDC. The ability to maintain integrity constraints makes our approach more effective than other approaches.

Image thresholding is one of the main techniques for image segmentation. It has many applications in pattern recognition, computer vision, and image and video understanding. This paper formulates the thresholding as an optimization problem: finding the best thresholds that minimize a weighted sum-of-squared-error function. A fast iterative optimization algorithm is presented to reach this goal. Our algorithm is compared with a classic, most commonly-used thresholding approach. Both theoretic analysis and experiments show that the two approaches are equivalent. However, our formulation of the problem allows us to develop a much more efficient algorithm, which has more applications, especially in real-time video surveillance and tracking systems.