GrOWL: A tool for visualization and editing of OWL ontologies

In an effort to optimize visualization and editing of OWL ontologies we have developed GrOWL—a browser and visual editor for OWL that accurately visualizes the underlying DL semantics of OWL ontologies while avoiding the difficulties of the verbose OWL syntax. In this paper, we discuss GrOWL visualization model and the essential visualization techniques implemented in GrOWL.     

从关系数据库学习OWL本体的方法

提出了一种从关系数据库半自动学习OWL本体的方法.该方法在形式化表示关系数据库模式和OWL本体的基础上,遵循从关系数据库模式到OWL本体的一组通用映射方法和规则,并基于Java 2平台实现了原型工具OntoLeamer.利用OntoLeamer进行的典型案例研究表明了该方法的有效性.     

W-KG2Vec: a weighted text-enhanced meta-path-based knowledge graph embedding for similarity search

Neural Computing and Applications - Recently, similar entity searching over knowledge graph (KG) has gained much attentions by researchers. However, in rich-semantic KGs with multi-typed entities...     

Extracting a justification for OWL ontologies by critical axioms

Extracting justifications for web ontology language (OWL) ontologies is an important mission in ontology engineering. In this paper, we focus on black-box techniques which are based on ontology reasoners. Through creating a recursive expansion procedure, all elements which are called critical axioms in the justification are explored one by one. In this detection procedure, an axiom selection function is used to avoid testing irrelevant axioms. In addition, an incremental reasoning procedure has been proposed in order to substitute series of standard reasoning tests w.r.t. satisfiability. It is implemented by employing a pseudo model to detect “obvious” satisfiability directly. The experimental results show that our proposed strategy for extracting justifications for OWL ontologies by adopting incremental expansion is superior to traditional Black-box methods in terms of efficiency and performance.     

Merging of axiomatic definitions of concepts in the complex OWL ontologies

In the last decade, ontology matching and mapping research has shown a measurable progress. This topic draws substantial attention within the research community, though it is not fully researched so far and new complex and effective solutions are needed. Current works are limited in finding alignments or mappings between concepts of heterogeneous ontologies. But, once ontology mappings are found, then how they (or their class expressions) are to be integrated automatically is left open for the ontology merging research. This paper elaborates the mapping of class expressions of concepts and contributes an algorithm for their merging in an automatic ontology merging process without any human intervention. However, the challenge of mapping axiomatic definitions is the most difficult task for merging concept definitions of the source ontologies, but it reveals significant increase in precision and recall values. In addition, with the study of these algorithms, we conclude that ontology merging facilitates when one wants to get ontology with the better quality as the combined rich axioms are added in the merged ontology. We also discuss the results of our first successful participation in the Conference, OA4QA and Anatomy track of OAEI 2015.     

Using OWL-VisMod through a decision-making process for reusing OWL ontologies

Knowledge representation (KR) can be defined as a set of ontological commitments, provided with the capabilities of performing inference. The knowledge can be represented using an ontology, which provides a shared insight into a certain domain. The use of ontologies to represent knowledge also allows interoperation among knowledge-based systems. The process of building ontologies can be tedious and sometimes exhaustive. A possible solution in order to avoid this problem would be to reuse the ontologies previously created by others. This paper describes a case study of reusability using OWL-VisMod, a tool designed for developing ontological engineering based on visual conceptual modelling for OWL ontologies. A workflow performed with OWL-VisMod is described; including a decision-making process in order to decide whether or not it could be desirable to reuse an ontology, according to the requirements of a certain project.     

Automatic extraction of OWL ontologies from UML class diagrams: a semantics-preserving approach

Full implementation of the Semantic Web requires widespread availability of OWL ontologies. Manual ontology development using current OWL editors remains a tedious and cumbersome task that requires significant understanding of the new ontology language and can easily result in a knowledge acquisition bottleneck. On the other hand, abundant domain knowledge has been specified by existing database schemata such as UML class diagrams. Thus developing an automatic tool for extracting OWL ontologies from existing UML class diagrams is helpful to Web ontology development. In this paper we propose an automatic, semantics-preserving approach for extracting OWL ontologies from existing UML class diagrams. This approach establishes a precise conceptual correspondence between UML and OWL through a semantics-preserving schema translation algorithm. The experiments with our implemented prototype tool, UML2OWL, show that the proposed approach is effective and a fully automatic ontology extraction is achievable. The proposed approach and tool will facilitate the development of Web ontologies and the realization of semantic interoperations between existing Web database applications and the Semantic Web.     

OWL 2: The next step for OWL

Since achieving W3C recommendation status in 2004, the Web Ontology Language (OWL) has been successfully applied to many problems in computer science. Practical experience with OWL has been quite positive in general; however, it has also revealed room for improvement in several areas. We systematically analyze the identified shortcomings of OWL, such as expressivity issues, problems with its syntaxes, and deficiencies in the definition of OWL species. Furthermore, we present an overview of OWL 2—an extension to and revision of OWL that is currently being developed within the W3C OWL Working Group. Many aspects of OWL have been thoroughly reengineered in OWL 2, thus producing a robust platform for future development of the language.     

Using OWL ontologies for adaptive patient information modelling and preoperative clinical decision support

We here present our research and experience regarding the design and implementation of a knowledge-based preoperative assessment decision support system. We discuss generic design considerations as well as the practical system implementation. We developed the system using semantic web technology, including modular ontologies developed in the OWL web ontology language, the OWL Java application programming interface and an automated logic reasoner. We discuss how the system enables to tailor patient information collection according to personalized medical context. The use of ontologies at the core of the system’s architecture permits to efficiently manage a vast repository of preoperative assessment domain knowledge, including classification of surgical procedures, classification of morbidities and guidelines for routine preoperative tests. Logical inference on the domain knowledge according to individual patient’s medical context enables personalized patients’ reports consisting of a risk assessment and clinical recommendations such as relevant preoperative tests.     

Combining OWL ontologies using -Connections

The standardization of the Web Ontology Language (OWL) leaves (at least) two crucial issues for Web-based ontologies unsatisfactorily resolved, namely how to represent and reason with multiple distinct, but linked ontologies, and how to enable effective knowledge reuse and sharing on the Semantic Web.In this paper, we present a solution for these fundamental problems based on -Connections. We aim to use -Connections to provide modelers with suitable means for developing Web ontologies in a modular way and to provide an alternative to the owl:imports construct.With such motivation, we present in this paper a syntactic and semantic extension of the Web Ontology language that covers -Connections of OWL-DL ontologies. We show how to use such an extension as an alternative to the owl:imports construct in many modeling situations. We investigate different combinations of the logics , and for which it is possible to design and implement reasoning algorithms, well-suited for optimization.Finally, we provide support for -Connections in both an ontology editor, SWOOP, and an OWL reasoner, Pellet.     

基于Word2Vec词嵌入和高维生物基因选择遗传算法的文本特征选择方法

文本特征是自然语言处理中的关键部分。针对目前文本特征的高维性和稀疏性问题,提出了一种基于Word2Vec词嵌入和高维生物基因选择遗传算法（GARBO）的文本特征选择方法,从而便于后续文本分类任务。首先,优化数据输入形式,使用Word2Vec词嵌入方法将文本转变成类似基因表示的词向量;然后,将高维词向量模拟基因表达方式进行迭代进化;最后,使用随机森林分类器对特征选择后的文本进行分类。使用中文评论数据集对所提出的方法进行实验,实验结果表明了优化后的GARBO特征选择方法在文本特征选择上的有效性,该方法成功地将300维特征降低为50维更有价值的特征,分类准确率达到88%,与其他过滤式文本特征选择方法相比,能够有效地降低文本特征维度,提高文本分类效果。     

Considerations about learning Word2Vec

The Journal of Supercomputing - Despite the large diffusion and use of embedding generated through Word2Vec, there are still many open questions about the reasons for its results and about its real...     

Patch2Vec: Globally Consistent Image Patch Representation

Many image editing applications rely on the analysis of image patches. In this paper, we present a method to analyze patches by embedding them to a vector space, in which the Euclidean distance reflects patch similarity. Inspired by Word2Vec, we term our approach Patch2Vec. However, there is a significant difference between words and patches. Words have a fairly small and well defined dictionary. Image patches, on the other hand, have no such dictionary and the number of different patch types is not well defined. The problem is aggravated by the fact that each patch might contain several objects and textures. Moreover, Patch2Vec should be universal because it must be able to map never‐seen‐before texture to the vector space. The mapping is learned by analyzing the distribution of all natural patches. We use Convolutional Neural Networks (CNN) to learn Patch2Vec. In particular, we train a CNN on labeled images with a triplet‐loss objective function. The trained network encodes a given patch to a 128D vector. Patch2Vec is evaluated visually, qualitatively, and quantitatively. We then use several variants of an interactive single‐click image segmentation algorithm to demonstrate the power of our method.