International Journal of Metadata, Semantics and Ontologies (8 papers in press)

• Semantic Information Extraction from Tamil Documents
  by Devakumar Jayaraman, Lakshmana Pandian subbiah, Geetha T.V. 
  Abstract: Semantic Information Extraction is a collective process of extracting concepts, entities, relations, entailment rules and important events from a document. We propose an approach towards gathering Semantic knowledge by extracting concepts, entities, and relations from the domain specific Tamil textual documents corpus. Concepts from the domain specific Tamil input documents will be extracted out by identifying semantic relationship between unique terms in documents with the help of lexical database such as WordNet. Entity extraction process identifies proper names in the given input document and then categorizes it into the predefined named entity categories. Relation between two entities or between two concepts or between concept and entity in each sentence are then extracted. This extracted semantic knowledge is exploited as a resource for extracting semantic information from the domain related documents.
  Keywords: Concept Extraction; Relation Extraction; Entity Extraction; Textual Entailment.
   
• Discovery and Evaluation of Non-Hierarchical Relations in Domain Ontologies
  by Albert Weichselbraun, Gerhard Wohlgenannt, Arno Scharl 
  Abstract: The identification and labeling of non-hierarchical relations are among the most challenging tasks in ontology learning. This paper describes a bottom-up approach for automatically suggesting ontology link types. The presented method extracts verb-vectors from semantic relations identified in the domain corpus, aggregates them by computing centroids for known relation types, and stores the centroids in a central knowledge base. Comparing verb-vectors extracted from unknown relations with the stored centroids yields link type suggestions. Domain experts evaluate these suggestions, refining the knowledge base and constantly improving the component's accuracy. Using four sample ontologies on ``energy sources'', this paper demonstrates how link type suggestion aids the ontology design process. It also provides a statistical analysis on the accuracy and average ranking performance of batch learning versus online learning.
  Keywords: ontology learning; ontology extension; link type detection; non-hierarchical relations; non-taxonomic relations; vector space model
   
• Ontology Engineering: Reuse and Integration
  by Vito Roberto, Elio Toppano, Rosalba Giuffrida, Giovanni B. Buora 
  Abstract: Reuse and integration are major steps in the ontology development process, often unavoidable to lower the costs of a new application. This is a position paper addressing the two issues and the problems encountered with an ontology-engineering approach. We provide a comprehensive account of the state of the art, and choose the domain of e-learning as a test-bed to verify and discuss the feasibility of methods and results. Practical issues have been addressed using the languages and tools for handling ontologies. We verify that reuse and integration of knowledge fragments are effective in constructing domain representations with larger coverage and enhanced reasoning capabilities, which significantly extend the expressiveness of semantic software layers. Problems emerge and concern the lack of explicit representations of the point of view and assumptions taken during the development: when reusing an ontology, one implicitly commits to the underlying conceptualization and associated context. We conclude that a new methodology should be put forward that views the development as a collaborative work within a community of stakeholders.
  Keywords: Ontologies; Ontology engineering; E-learning; Methodologies of software engineering.
   
• Structured storage of legal precedents using a minimal deontic ontology, for computer assisted legal document querying
  by Alan Abrahams, David Eyers 
  Abstract: The automatic interpretation of legal precedents, or case law, by computerized natural language processing algorithms remains an elusive challenge. This paper proposes a less ambitious goal: human assisted semantic tagging of case law, using a basic deontic ontology, so that the structured document can be queried. The minimal encoding presented is sufficient to represent a comprehensive range of deontic concepts. Independent RDF named graphs are used to represent the numerous conflicting perspectives on the situation being examined within a legal case. An implementation of event calculus is used to make inference over these named graphs. To demonstrate that useful semantic searching can be performed, we employ our minimal deontic encoding to encode the significant aspects of a commercial case from the South African High Court.
  Keywords: deontic; legal document querying; semantic tags; RDF named graphs
   
• A Scenario for the development and use of Teaching-Oriented Ontologies
  by Vasilios S. Belesiotis, Nikolaos Alexandris 
  Abstract: In this paper we propose a methodology for the construction and use of Teaching-Oriented Ontologies and present how such ontologies can support the lesson planning phase and the teaching process. The presented approach develops Teaching-Oriented Ontologies with respect to the students’ educational level and follows the relevant textbooks and the aims and goals of the appropriate curriculum. We present its pilot use with focus on the Greek secondary-education system. We then present our arguments supporting the use of such ontologies as tools for organizing subjects, supporting the teaching process, and enhancing the evaluation procedure. In addition, we suggest a methodology for the development of such ontologies and explain how teaching-oriented ontologies can aid the teaching process. Finally, we demonstrate an example of the use of our methodology and discuss the findings of an empirical evaluation of its use by teachers during the lesson planning phase.
  Keywords: didactics; teaching; education; teaching-oriented ontologies; ICT; knowledge representation; intelligent scenario systems
   
• GVP Model based Temporal Visualization of User Centric Data
  by T.Mala Nehru 
  Abstract: Information visualization produces visual representations of abstract data to reinforce human cognition and perception. Information that can be transformed to visual form is of various types and here it is proposed to develop a system that performs temporal visualization of a user’s personal interest information which is liable to vary over time. The system performs visualization of the user’s profile recorded automatically from his browsing interest and visualizes the search results of a search query which is a temporal data by applying the identified personalized user profile. The basis of comparison is automatically created ontologies indicating domain ontology and user ontology. Thus the system performs a user centric temporal visualization based on a novel Gestalt Visual Perception Model as it focuses on a user’s personal interest that changes over a period of time. The system automatically constructs the user profile from a domain ontology which is constructed automatically using a statistical technique called as Latent Semantic Indexing and also from the user’s browsing history. The Gestalt visual perception model defines the mapping functions in transforming the personalized data into visual scenes and then deriving the tasks out of the visual scene. These visualizations helps in getting the insight from the data visualized.
  Keywords: Ontology visualization, Gestalt perception principles, Personalized search, User centric temporal visualization.
   
• Ontological Technologies for User Modeling
  by Sergey Sosnovsky, Darina Dicheva 
  Abstract: This paper brings together research from two different fields: user modeling and Web-ontologies – in attempt to demonstrate how recent semantic trends in Web development can be combined with the modern technologies of user modeling. Over the last several years, a number of user-adaptive systems have been exploiting ontologies for the purposes of semantics representation, automatic knowledge acquisition, domain and user model visualization and creation of interoperable and reusable architectural solutions. Before discussing these projects we first overview the underlying user modeling and ontological technologies. As an example of the project employing ontology-based user modeling, we present an experiment design for translation of overlay student models for relative domains by means of ontology mapping.
  Keywords: user modeling, ontologies, adaptation, semantic web
   
• SSERank: Semantic Search Engine for Page Ranking Based on the Relations Weight
  by Thabet Slimani, Boutheina Ben Yaghlane, Khaled Mellouli 
  Abstract: With the massive growth of information in the Web, semantic search engines promise to provide more accurate results than current keywords search engines. However, progress with semantic search has been delayed due to the limitation of exploiting an important content of Semantic Web resources, which are, relations (semantic relations and semantic associations). In this paper, we explore a novel approach (SSERank) of adapting keywords to query the semantic web: the approach automatically exploits the keywords in a user query to find related concepts in the ontology. Given an ontology and a knowledge base of annotated pages, weights are assigned to relations based on some properties of the ontology, so that they measure the strength of the relation. However, in order to rank results, most of the existing solutions need to give a weight for each page in the whole annotated knowledge base. Using this model, we have developed two algorithms for ranking the importance of web pages: A direct relation-based algorithm and a semantic association-based algorithm (indirect relation). This paper proposes a prototype relation-based page weighting applied to a Semantic Web search engine synthetically constructed in our Lab. The obtained results show that SSERank will have great utility although there is potential for improvement.
  Keywords: Semantic Web; semantic search; relation weight; page rank; semanticrelation; semantic association; knowledge retrieval