Mining the Semantic Web

In the Semantic Web vision of the World Wide Web, content will not only be accessible to humans but will also be available in machine interpretable form as ontological knowledge bases. Ontological knowledge bases enable formal querying and reasoning and, consequently, a main research focus has been the investigation of how deductive reasoning can be utilized in ontological representations to enable more advanced applications. However, purely logic methods have not yet proven to be very effective for several reasons: First, there still is the unsolved problem of scalability of reasoning to Web scale. Second, logical reasoning has problems with uncertain information, which is abundant on Semantic Web data due to its distributed and heterogeneous nature. Third, the construction of ontological knowledge bases suitable for advanced reasoning techniques is complex, which ultimately results in a lack of such expressive real-world data sets with large amounts of instance data. From another perspective, the more expressive structured representations open up new opportunities for data mining, knowledge extraction and machine learning techniques. If moving towards the idea that part of the knowledge already lies in the data, inductive methods appear promising, in particular since inductive methods can inherently handle noisy, inconsistent, uncertain and missing data. While there has been broad coverage of inducing concept structures from less structured sources (text, Web pages), like in ontology learning, given the problems mentioned above, we focus on new methods for dealing with Semantic Web knowledge bases, relying on statistical inference on their standard representations. We argue that machine learning research has to offer a wide variety of methods applicable to different expressivity levels of Semantic Web knowledge bases: ranging from weakly expressive but widely available knowledge bases in RDF to highly expressive first-order knowledge bases, this paper surveys statistical approaches to mining the Semantic Web. We specifically cover similarity and distance-based methods, kernel machines, multivariate prediction models, relational graphical models and first-order probabilistic learning approaches and discuss their applicability to Semantic Web representations. Finally we present selected experiments which were conducted on Semantic Web mining tasks for some of the algorithms presented before. This is intended to show the breadth and general potential of this exiting new research and application area for data mining.     

An ontology-based knowledge framework for engineering material selection

Engineering material selection intensively depends on domain knowledge. In the face of the large number and wide variety of engineering materials, it is very necessary to research and develop an open, shared, and scalable knowledge framework for implementing domain-oriented and knowledge-based material selection. In this paper, the fundamental concepts and relationships involved in all aspects of material selection are analyzed in detail. A novel ontology-based knowledge framework is presented. The ontology-based Semantic Web technology is introduced into the semantic representation of material selection knowledge. The implicit material selection knowledge is represented as a set of labeled instances and RDF instance graphs in terms of the concept model, which provides a formal approach to organizing the captured material selection knowledge. A knowledge retrieval and reasoning approach integrating ontology concepts, instances, knowledge rules, and semantic queries encoded with Query-enhanced Web Rule Language (SQWRL) is proposed. The presented knowledge framework can provide powerful knowledge services for material selection. Finally, based on this knowledge framework, a case study on constructing a mold material selection knowledge system is provided. This work is a new attempt to build an open and shared knowledge framework for engineering material selection.     

Making biomedical ontologies and ontology repositories work

It is becoming impossible to contemplate successful bio-medical research without canonical data structures. The biomedical computation community finds itself grappling with hundreds of different knowledge bases, metadata formats, and database schemas. These include primary databases, such as those in GenBank and MEDLINE; metadata that describe the primary data, such as those in caBIO; and knowledge bases that codify biomedical concepts, such as the Gene Ontology and SNOMED-CT. These data structures are representable in languages such as DICOM and MAGE-ML. Many of these data elements and knowledge bases have emerged out of necessity from work that scientists, unfamiliar with data and knowledge representation standards, have done in isolation. Many of these resources fail to follow consistent modeling conventions, so computer programs cannot consistently interpret them. Semantic Web technology and languages such as RDF and OWL can rectify the problem somewhat by providing a common metadata and ontology language and Web-based tools for dealing with ontologies and knowledge structures. However, even if translation mechanisms exist between various biomedical resources and Semantic Web languages (which, by itself, is unlikely to happen for all resources), this translation is only part of the solution.     

Revyu: Linking reviews and ratings into the Web of Data

Revyu is a live, publicly accessible reviewing and rating Web site, designed to be usable by humans whilst transparently generating machine-readable RDF metadata for the Semantic Web, based on user input. The site uses Semantic Web specifications such as RDF and SPARQL, and the latest Linked Data best practices to create a major node in a potentially Web-wide ecosystem of reviews and related data. Throughout the implementation of Revyu design decisions have been made that aim to minimize the burden on users, by maximizing the reuse of external data sources, and allowing less structured human input (in the form of Web 2.0-style tagging) from which stronger semantics can later be derived. Links to external sources such as DBpedia are exploited to create human-oriented mashups at the HTML level, whilst links are also made in RDF to ensure Revyu plays a first class role in the blossoming Web of Data. In this paper we document design decisions made during the implementation of Revyu, discuss the techniques used for linking Revyu data with external sources, and outline how data from the site is being used to infer the trustworthiness of reviewers as sources of information and recommendations.     

面向知识网格的本体学习研究

网格计算正在从单纯的面向大型计算的分布式资源共享发展为一种面向服务的架构,以实现透明而可靠的分布式系统集成。网格智能是指如何获取、预处理、表示和集成不同层次的网格服务(如HTML/XML/RDF/OWL文档、服务响应时间和服务质量等)的数据和信息,并最终转换为有用的智能(知识)。因为高层知识将在未来的网格应用起到越来越重要的作用,本体是知识网格实现的关键。文章提出了一种实现从Web文档中本体(半)自动构建的本体学习框架WebOntLearn,并讨论了本体学习中领域概念的抽取、概念之间关系的抽取和分类体系的自动构建等关键技术。     

Automated ontology construction for unstructured text documents

Ontology is playing an increasingly important role in knowledge management and the Semantic Web. This study presents a novel episode-based ontology construction mechanism to extract domain ontology from unstructured text documents. Additionally, fuzzy numbers for conceptual similarity computing are presented for concept clustering and taxonomic relation definitions. Moreover, concept attributes and operations can be extracted from episodes to construct a domain ontology, while non-taxonomic relations can be generated from episodes. The fuzzy inference mechanism is also applied to obtain new instances for ontology learning. Experimental results show that the proposed approach can effectively construct a Chinese domain ontology from unstructured text documents.     

Towards Ontology Generation from Tables

At the heart of today's information-explosion problems are issues involving semantics, mutual understanding, concept matching, and interoperability. Ontologies and the Semantic Web are offered as a potential solution, but creating ontologies for real-world knowledge is nontrivial. If we could automate the process, we could significantly improve our chances of making the Semantic Web a reality. While understanding natural language is difficult, tables and other structured information make it easier to interpret new items and relations. In this paper we introduce an approach to generating ontologies based on table analysis. We thus call our approach TANGO (Table ANalysis for Generating Ontologies). Based on conceptual modeling extraction techniques, TANGO attempts to (i) understand a table's structure and conceptual content; (ii) discover the constraints that hold between concepts extracted from the table; (iii) match the recognized concepts with ones from a more general specification of related concepts; and (iv) merge the resulting structure with other similar knowledge representations. TANGO is thus a formalized method of processing the format and content of tables that can serve to incrementally build a relevant reusable conceptual ontology.     

Semantic Web Constraint Language and its application to an intelligent shopping agent

Semantic Web society was initially focused only on data, but then gradually moved toward knowledge. If a vision of the Semantic Web is to enhance humans' decision-making assisted by machines, a missing but important part is knowledge about constraints on data and concepts represented by ontology. This paper proposes a Semantic Web Constraint Language (SWCL) based on OWL, and shows its effectiveness in representing and solving an internet shopper's decision-making problems by implementing a shopping agent in the Semantic Web environment.     

软件工程关联数据的自动构建

针对目前在分布异构的大规模软件开发中难以高效地知晓信息和发现知识的问题,将语义网引入软件工程领域,对多源异构数据进行细粒度语义关联,提出本体构建、关联抽取和发现的方法,实现基于本体的软件工程关联数据的自动构建。该方法对软件工程本体进行概念抽取、合并、实例消解和属性消歧,从软件仓库结构化数据集中抽取出完整无冗余的关联数据;并采用同义词、动宾短语和结构关系三个特征利用自然语言处理(NLP)技术和信息检索(IR)技术从软件仓库中发现潜在的关联数据。实验结果表明,所提出的方法能从分布式软件工程数据集中自动构建和融合生成软件工程本体,并有效地发现潜在的关联数据将其扩充到软件工程本体中;与Baseline、Phraing和O-CSTI三种方法相比,关联数据发现的召回率、精准率和F值都有显著提高。     

A redundancy-based method for the extraction of relation instances from the Web

The Semantic Web requires automatic ontology population methods. We developed an approach, that given existing ontologies, extracts instances of ontology relations, a specific subtask of ontology population. We use generic, domain-independent techniques to extract candidate relation instances from the Web and exploit the redundancy of information on the Web to compensate for loss of precision caused by the use of these generic methods. The candidate relation instances are then ranked based on co-occurrence with a small seed set. In an experiment, we extracted instances of the relation between artists and art styles. The results were manually evaluated against selected art resources. The method was also tested in the football domain. We also compare the performance of our ranking to that of a Google-hit count-based method.     

A high-level electrical energy ontology with weighted attributes

One of the significant application areas of domain ontologies is known to be text analysis applications like information extraction and text classification systems, and semantic portals. In this paper, we present a high-level ontology for the electrical energy domain. This domain ontology has weighted attributes to cover the inherent fuzziness in the textual representations of its concepts. Additionally, we have included in the ontology the necessary attributes to align the ontology concepts to on-line collaborative knowledge bases like Wikipedia and linked open data sources like DBpedia, other attributes to facilitate its use in multilingual applications, and concepts to hold the named entities in the domain. The ultimate ontology is aligned with the previously proposed ontologies for the energy-related subdomains after extending the latter ones with weighted attributes. We make the ultimate form of the electrical energy ontology, as well as the extended versions of the domain ontologies for the subdomains, available for research purposes. Also included in the paper are sample text analysis applications which mainly exploit the weighted attributes within the ontology.     

RDF数据浏览的研究综述

随着语义网的快速发展,目前Web上语义网数据已经达到相当的规模,成为重要的信息和知识来源.因此,RDF数据浏览的研究开始得到广泛关注.通过对比传统Web信息浏览和RDF数据浏览两个问题,指出RDF数据浏览的5个重要问题:确定浏览子图的模式、数据的收集、大规模数据的处理、数据的组织方式以及数据的呈现方式.基于这些挑战,我们调研了多个系统和不同的解决方案.最后,总结了目前的研究现况,讨论存在的挑战,并提出未来的研究方向.