Name ambiguity refers to a problem that different people might be referenced with an identical name. This problem has become critical in many applications, particularly in online bibliography systems, such as DBLP and CiterSeer. Although much work has been conducted to address this problem, there still exist many challenges. In this paper, a general framework of constraint-based topic modeling is proposed, which can make use of user-defined constraints to enhance the performance of name disambiguation. A Gibbs sampling algorithm that integrates the constraints has been proposed to do the inference of the topic model. Experimental results on a real-world dataset show that significant improvements can be obtained by taking the proposed approach.

In order to manage the increasing amount of RDF data, an RDF repository should provide not only necessary scalability and efficiency, but also sufficient inference capabilities. Though existing RDF repositories have made progress towards this goal, there is still ample space for improving the overall performance. In this paper, we propose a native RDF repository, System II, to pursue a better tradeoff among the system scalability, the query efficiency, and the inference capabilities. System II takes the hypergraph representation for RDF as the data model for its persistent storage, which effectively avoids the costs of data model transformation when accessing RDF data. Based on this native storage scheme, a set of efficient semantic query processing techniques are designed. First, several indices are built to accelerate RDF data access including a value index, a labeling scheme for transitive closure computation, and three triple indices. Second, we propose a hybrid inference strategy under the pD* semantics to support inference for OWL-Lite with a relatively low computational complexity. Finally, we extend the SPARQL algebra to explicitly express inference semantics in logical query plan by defining new algebra operators. The results of performance evaluation on the LUBM benchmark show that System II has a better combined metric value than the other comparable systems.

This paper presents an algorithm to improve the performance of item-based collaborative filtering algorithms working with sparse data sets. The factors impacting the correlation calculation in item-based collaborative filtering algorithms were analyzed to develop an item relationship density as an important characteristic for describing the rating matrix, the effect of the item relationship density on item-based collaborative filtering is then illustrated. The item relation density is then used to develop a virtual user filling algorithm. That effectively improves the precision and coverage of item-based algorithms with sparse datasets. Thus the item relation density is a key characteristic factor for rating matrices.

Most collaborative filtering(CF) research has focused on doing experiments on single dataset or datasets with the same characteristics.This paper presents an analysis of several typical CF algorithms,the User-based KNN method(with 20 neighborhoods),the item-based method,the item average method,the item user average method,and the Slope One method.These algorithms are evaluated on two types of datasets,Movielens and Book-Crossing,which have different user-item distribution characteristics.The results show for the relatively dense ratings on the Movielens dataset,the Slope One method has the best prediction precision,while on datasets with relatively sparse ratings such as Book-Crossing,the item-based method is the best,while the Slope One method is the worst.Thus,the different CF algorithms give different results on the different datasets,so the CF algorithm should be designed according to the user-item distribution characters.

Data correlations are relationships among I/O documents of Web services. They reflect the correlations among services on business logics, and are important to service tasks such as service composition, service discovery, and etc. Existing modeling methods of data correlation express data correlations indirectly by reasoning on data models or ontology models used by service interface definitions, and they have limitations on expressiveness. In order to solve this problem, referring to the idea of hyperlinks, this paper proposes the concept of Explicit Data-Link, which expresses data correlations as static explicit declarations. It first gives the abstract model of explicit data-link and its XML implementation. Then, the application of explicit data-link is discussed in detail. A study case that makes use of explicit data-link to optimize the composition algorithm for WS-Challenge 2007 is given. The experiment results show that this optimization method is effective when dealing with data sets that have complex inheritance structures, and the algorithm wins the championship of performance in the competition.

To manage the increasing amount of RDF data, an RDF repository should provide not only necessary scalability and efficiency, but also sufficient inference capabilities. In this paper, we propose a native RDF repository, System, to pursue a better tradeoff among the above requirements. System takes the hypergraph representation for RDF as the data model for its persistent storage, which effectively avoids the costs of data model transformation when accessing RDF data. In addition, a set of efficient semantic query processing techniques are designed. The results of performance evaluation on the LUBM benchmark show that System has a better combined metric value than the other comparable systems.

Name ambiguity is a critical problem in many applications, in particular in the online bibliography systems, such as DBLP and CiteSeer. Previously, several clustering based methods have been proposed although, the problem still presents to be a big challenge for both research and industry communities. In this paper, we present a complementary study to the problem from another point of view. We propose an approach of finding atomic clusters to improve the performance of existing clustering-based methods. We conducted experiments on a dataset from a real-world system: Arnetminer.org. Experiments results show that significant improvements can be obtained by using the proposed atomic clusters finding approach (about +8% and +27% improvements depending on different clustering methods).

New event detection (NED) is aimed at detecting from one or multiple streams of news stories the one being reported on a new event (i.e.not reported previously).Preliminary experiments show that terms of different types (e.g.Noun and Verb) have different effects for different classes of stories in determining whether or not two stories are on the same topic.Unfortunately, conventional approaches usually ignore the fact.This paper proposes a NED model utilizing two approaches to addressing the problem based on term reweighting.In the first approach, the paper proposes to employ statistics on training data to learn the model for each class of stories, and in the second, the paper proposes to adjust term weights dynamically based on previous story clusters.Experimental results on two linguistic data consortium (LDC) data sets: TDT2 and TDT3 show that both the proposed approaches can effectively improve the performance of NED task, compared to the baseline method and existing methods.

In order to solve the conformity of personalized services with legacy systems, we design and implement a personalized digital library framework based on Service Oriented Architecture (SOA). In this paper, we introduce the framework development as following: Analyzing current systems and their workflow with SOA design pattern, and extracting needed personalized service interface; implement the Enterprise components and data interface of those service interface according to SOA standard; composing the service components according to workflow, so that to build up the whole framework.

Most previous research focus on organizing news set into flat collections of stories. However, a topic in news is more than a mere collection of stories: it is characterized by a definite structure of inter-related events. Stories within a topic usually share some terms which are related to the topic other than a specific event, so stories of different events are usually very similar to each other within a topic. To deal with this problem, we propose a new event identification method based on the term committee. We first capture some tight term clusters as term committees of potential events, and then use them to re-weight the key terms in a story. The experimental results on two Linguistic Data Consortium (LDC) datasets show that the proposed method for event identification outperforms previous methods significantly.

More and more ontologies have been published and used widely on the web. In order to make good use of an ontology, especially a new and complex ontology, we need methods to help understand it first. Identifying potentially important concepts and relations in an ontology is an intuitive but challenging method. In this paper, we first define four features for potentially important concepts and relation from the ontological structural point of view. Then a simple yet effective Concept-And-Relation-Ranking (CARRank ) algorithm is proposed to simultaneously rank the importance of concepts and relations. Different from the traditional ranking methods, the importance of concepts and the weights of relations reinforce one another in CARRank in an iterative manner. Such an iterative process is proved to be convergent both in principle and by experiments. Our experimental results show that CARRank has a similar convergent speed as the PageRank-like algorithms, but a more reasonable ranking result.

To manage the increasing amount of RDF data, an RDF repository should provide not only necessary scalability and efficiency, but also sufficient inference capabilities. In this paper, we propose a native RDF repository, System, to pursue a better tradeoff among the above requirements. System takes the hypergraph representation for RDF as the data model for its persistent storage, which effectively avoids the costs of data model transformation when accessing RDF data. In addition, a set of efficient semantic query processing techniques are designed. The results of performance evaluation on the LUBM benchmark show that System has a better combined metric value than the other comparable systems.

Ranking the importance of concepts and the weights of relations is an effective method for evaluating an ontology, which can improve the design of ontology for domain expert, and be used as a component of semantic Web search engine. Current link analysis ranking algorithms cannot be directly applied to rank concepts, or efficiently to assign weights to relations. According to the characteristic of ontology graph structure, an algorithm is proposed with Hub rating instead of Authority rating as importance of concepts. The algorithm mutually reinforces importance of concepts and weights of relations in the iteration process which is proved to converge to the fixpoint of equations. The experimental results show the algorithm has the similar convergence speed to PageRank but more reasonable ranking of concepts importance and relations weights.

This paper is concerned with the problem of name disambiguation. By name disambiguation, we mean distinguishing persons with the same name. It is a critical problem in many knowledge management applications. Despite much research work has been conducted, the problem is still not resolved and becomes even more serious, in particular with the popularity of Web 2.0. Previously, name disambiguation was often undertaken in either a supervised or unsupervised fashion. This paper first gives a constraint-based probabilistic model for semi-supervised name disambiguation. Specifically, we focus on investigating the problem in an academic researcher social network (http://arnetminer.org). The framework combines constraints and Euclidean distance learning, and allows the user to refine the disambiguation results. Experimental results on the researcher social network show that the proposed framework significantly outperforms the baseline method using unsupervised hierarchical clustering algorithm.

To solve the problem of lacking effective services query mechanism in existing Web services and maximize the using of Web services,a semantic Web technology based Web services query mechanism was proposed.With this mechanism,the Web services related ontology information,domain dependent and independent basic conceptual ontology information were adopted.Combined with a former system Web Oriented Distributed Ontology System(WODOS) which supported ontology-based reasoning, a new system Web Services Intelligent Discovery System(WSIDS) based on domain rules and F-logic reasoning engine was implemented.At the same time,Web services constraint model: Subject,Property,Value,Action(SPVA),and semantic supplement model were proposed.Finally,an application case was used to illustrate reasoning process and running result of this system,and was also compared with other related tasks.

Because of the difference in computing ability of various mobile devices,there is no enough good system that converts current Web contents into the different size according to the mobile device’s ability.This article introduces a Web content adaptation system based on Web annotation that can generate small pages proper to mobile device’s ability and does not need any changes about original Web.Compared to other adaptation system this system can improve some functions: More proper adaptation to different mobile devices and converting speed and so on.

Web服务中消息的异构性使服务发现准确率降低并且它对组装后Web服务流程的运行有很大的影响,对此,提出了一种基于本体的Web服务异构消息匹配及其重用机制,通过概念抽取、合并及匹配组合等方法将已有匹配结果应用到下一次匹配过程中,使得消息自动转换系统在与领域和流程相关的匹配能力上有了明显增强．实验表明,匹配的准确率和召回率平均提高了11．1％和13．2％．