| |||
|
User-Centred Onlology Learning for Knowledge Management User-Centred Onlology Learning for Knowledge Management Brewster, Mr. Christopher and Ciravegna, Dr. Fabio and Wilks, Prof. Yorick (2002) User-Centred Onlology Learning for Knowledge Management. In Proceedings 7th International Workshop on Applications of Natural Language to Information Systems, Stockholm. Full text available as: PDF - Requires Adobe Acrobat Reader or other PDF viewer. Automatic ontology building is a vital issue in many fields where they are currently built manually. This paper presents a user-centred methodology for ontology construction based on the use of Machine Learning and Natural Language Processing. In our approach, the user selects a corpus of texts and sketches a preliminary ontology (or selects an existing one) for a domain with a preliminary vocabulary associated to the elements in the ontology (lexicalisa-tions). Examples of sentences involving such lexicalisation (e.g. ISA relation) in the corpus are automatically retrieved by the system. Retrieved examples are validated by the user and used by an adaptive Information Extraction system to generate patterns that discover other lexicalisations of the same objects in the ontology, possibly identifying new concepts or relations. New instances are added to the existing ontology or used to tune it. This process is repeated until a satisfactory ontology is obtained. The methodology largely automates the on-tology construction process and the output is an ontology with an associated trained leaner to be used for further ontology modifications. Keywords: Knowledge management, natural language processing, ontology learning, document annotation Subjects: AKT Challenges > Knowledge acquisition ID Code: 125 Deposited By: Brewster, Christopher Deposited On: 27 February 2003 Alternative Locations: http://www.dcs.shef.ac.uk/~fabio/cira-papers.html Wiki Whiteboard: User_Centred_Onlology_Learning_for_Knowledge_Management (requires valid AKT password) (experimental) D3Eprints discussion: View public discussion of this document Create private discussion of this document Edit this record (Requires valid AKT password) Contact the site administrator at: hg@ecs.soton.ac.uk Combining Rule-Based Methods and Latent Semantic Analysis for Ontology Structure ... Combining Rule-Based Methods and Latent Semantic Analysis for Ontology Structure Construction Guergana K. Savovaa, PhD, Dara Beckera, Marcelline Harrisa,b, R.N., PhD and Christopher G. Chutea, MD, DrPH aDivision of Medical Informatics Research, Mayo Clinic, Rochester, MN bDivision of Nursing Research, Department of Nursing, Mayo Clinic, Rochester, MN Abstract The goal of this research is to identify automated and semi-automated techniques for organizing terms into an ontology for functioning, disability, and health. Such an ontology can be built by importing terms from existing classifications and by adding to these, terms extracted from everyday practice medical language, e.g., medical reports. Our research extends previous work by combining a rule-based technique and an unsupervised machine-learning method to semantically clus-ter terms from medical reports. The rule-based approach identifies lexico-syntactic patterns suggestive of parent-child relations and coordination structures suggestive of sibling-sibling relations. Latent semantic analysis (LSA) is applied to those relations to weed spurious pairs. Our methodology is corpus-based. The evaluation is done by two domain experts. Applying LSA as a filter to the relations extracted by rules did not improve the precision for either parent-child relations (52%) or sibling-sibling relations (71%). Future research includes bootstrapping techniques to grow the semantic clus-ters, which is believed to show the advantages of using LSA. The significance of this work is its application in ontology construction, thesaurus expansion, and information retrieval. Keywords: Natural language processing, Machine learning, Terminology, Medical informatics, Classification. Introduction This paper is a report of work on the development of an on-tology for functioning, disability, and health at Mayo Clinic. Natural language processing (NLP) techniques offer solutions to the challenges in coding this domain content. The work in this paper ... DODDLE-OWL: On-the-fly Ontology Construction ... DODDLE-OWL: On-the-fly Ontology Construction with Ontology Quality Management Takeshi Morita��, Yoshihiro Shigeta��, Naoki Sugiura��, Naoki Fukuta��, Noriaki Izumi, and Takahira Yamaguchi ��Shizuoka University, 3-5-1 Johoku, Hamamatsu, Shizuoka 432-8011, Japan National Institute of AIST, 2-41-6, Aomi, Koto-ku, Tokyo, Japan Keio University, 4-1-1 Hiyoshi, Kohoku-ku, Yokohama-shi, Kanagawa, Japan morita@ks.cs.inf.shizuoka.ac.jp, yamaguti@ae.keio.ac.jp 1 Introduction In this paper, we propose a software environment for user-centered on-the-fly ontology construction named DODDLE-OWL (Domain Ontology rapiD DeveLopment Environment - Web Ontology Language extension). The architecture of DODDLE-OWL is re-designed based on DODDLE-II [1], our former study. DODDLE-OWL has the following five modules: Input Module, Construction Module, Refinement Module, Visualization Module, and Translation Module. DODDLE-OWL supports the construction of both taxonomic relationships and nontaxonomic relationships in ontologies. Since DODDLE-II has been built for ontology construction not for the Semantic Web but for typical knowledge systems, it needs some extensions for the SemanticWeb such as OWL export facility. DODDLE-OWL contributes the evolution of ontology construction and the Semantic Web. 2 The DODDLE-OWL Architecture Figure 1 shows the o ... |
